# Publication No.

FHWA NHI 01-004 December 2001

Hydraulic Design Series Number 6

RIVER ENGINEERING FOR HIGHWAY ENCROACHMENTS

Highways in the River Environment

Technical Report Documentation Page

1.

Report No.

FHWA NHI 01-004 HDS 6

2.

Government Accession No.

3.

Recipient's Catalog No.

4.

Title and Subtitle RIVER ENGINEERING FOR HIGHWAY ENCROACHMENTS Highways in the River Environment

5. 6. 8.

Report Date December 2001 Performing Organization Code Performing Organization Report No.

7.

Author(s) E.V. Richardson, D.B. Simons, P.F. Lagasse

9.

Performing Organization Name and Address Ayres Associates 3665 JFK Parkway Building 2, Suite 200 Fort Collins, Colorado 80525

10. Work Unit No. (TRAIS)

11. Contract or Grant No. DTFH61-99-T-25051

12. Sponsoring Agency Name and Address Office of Bridge Technology FHWA, Room 3203 400 Seventh Street, SW Washington, D.C. 20590 15. Supplementary Notes National Highway Institute 4600 North Fairfax Dr., Suite 800 Arlington, Virginia 22203

13. Type of Report and Period Covered

14. Sponsoring Agency Code

Project Manager: Philip L. Thompson Technical Assistants: Larry Arneson, J. Sterling Jones, Jorge E. Pagán-Ortiz, FHWA; P.E. Clopper, J.D. Schall, S.A. Schumm, L.W. Zevenbergen, Ayres Associates The Federal Highway Administration document "Highways in the River Environment - Hydraulic and Environmental Design Considerations" was first published in 1975, was revised in 1990, and is now issued as Hydraulic Design Series 6, "River Engineering for Highway Encroachments." This document has proven to be a singularly authoritative document for the design of highway associated hydraulic structures in moveable boundary waterways. This revised document incorporates many technical advances that have been made in this discipline since 1990. In addition, Hydraulic Engineering Circulars (HEC) 18, 20, and 23, have been published since 1990. This document and the HECs provide detailed guidance on stream instability, scour, and appropriate countermeasures. In HDS-6, hydraulic problems at stream crossings are described in detail and the hydraulic principles of rigid and moveable boundary channels are discussed. In the United States, the average annual damage related to hydraulic problems at highway facilities on the Federal-aid system is $40 million. Damages by streams can be reduced significantly by considering channel stability. The types of river changes to be carefully considered relate to: (1) lateral bank erosion; (2) degradation and aggradation of the streambed that continues over a period of years, and (3) natural short-term fluctuations of streambed elevation that are usually associated with the passage of floods. The major topics are: sediment transport, natural and human induced causes of waterway response, stream stabilization (bed and banks), hydraulic modeling and computer applications, and countermeasures. Case histories of typical human and natural impacts on waterways are analyzed. 17. Key Words aggradation, degradation, alluvial channel, alluvial fan, river training, geomorphology, headcutting, lateral migration, riprap, sediment transport, scour, stable channel design, countermeasures 19. Security Classif. (of this report) Unclassified 18. Distribution Statement This document is available to the public through the National Technical Information Service, Springfield, VA 22161 (703) 487-4650 and isddc.dot.gov 21. No. of Pages 644 22. Price

20. Security Classif. (of this page) Unclassified

Form DOT F 1700.7 (8-72)

Reproduction of completed page authorized

TABLE OF CONTENTS

LIST OF FIGURES ............................................................................................................ xiii LIST OF TABLES ............................................................................................................. xxv ACKNOWLEDGMENTS .................................................................................................. xxix SYMBOLS ....................................................................................................................... xxxi GLOSSARY................................................................................................................... xxxvii 1. INTRODUCTION ...................................................................................................... 1.1 1.1 CLASSIFICATION OF RIVER CROSSINGS AND ENCROACHMENTS ......... 1.1 1.1.1 1.1.2 1.2 Types of Encroachment...................................................................... 1.2 Geometry of Bridge Crossings............................................................ 1.2

DYNAMICS OF NATURAL RIVERS AND THEIR TRIBUTARIES.................... 1.2 1.2.1 1.2.2 Historical Evidence of the Natural Instability of Fluvial Systems.......... 1.3 Introduction to River Hydraulics and River Response ......................... 1.7

1.3

EFFECTS OF HIGHWAY CONSTRUCTION ON RIVER SYSTEMS............... 1.9 1.3.1 1.3.2 Immediate Response of Rivers to Encroachment ............................... 1.9 Delayed Response of Rivers to Encroachment..................................1.10

1.4 1.5

EFFECTS OF RIVER DEVELOPMENT ON HIGHWAY ENCROACHMENTS......................................................................................1.11 TECHNICAL ASPECTS..................................................................................1.13 1.5.1 1.5.2 1.5.3 Variables Affecting River Behavior.....................................................1.14 Basic Knowledge Required................................................................1.14 Data Requirements............................................................................1.14

1.6

FUTURE TECHNICAL TRENDS ....................................................................1.15 1.6.1 1.6.2 1.6.3 Adequacy of Current Knowledge .......................................................1.15 Research Needs ................................................................................1.16 Training .............................................................................................1.17

1.7

OVERVIEW OF MANUAL CONTENTS ..........................................................1.17 1.7.1 1.7.2 1.7.3 1.7.4 1.7.5 1.7.6 1.7.7 Chapter 2 - Open Channel Flow ........................................................1.17 Chapter 3 - Fundamentals of Alluvial Channel Flow...........................1.18 Chapter 4 - Sediment Transport ........................................................1.18 Chapter 5 - River Morphology and River Response ...........................1.19 Chapter 6 - River Stabilization and Bank Protection .........................1.19 Chapter 7 - Scour at Bridges .............................................................1.19 Chapter 8 - Data Need and Data Sources .........................................1.20

i

1.7.8

Chapter 9 - Design Considerations for Highway Encroachment and River Crossings .......................................................1.20 1.7.9 Chapter 10 - Overview Examples of Design for Highways in the River Environment ........................................................1.21 1.7.10 Chapter 11 - References ...................................................................1.21 1.7.11 Appendices........................................................................................1.21 1.8 2. DUAL SYSTEM OF UNITS.............................................................................1.21

OPEN CHANNEL FLOW ........................................................................................... 2.1 2.1 INTRODUCTION ............................................................................................. 2.1 2.1.1 2.2 Definitions........................................................................................... 2.1

THREE BASIC EQUATIONS........................................................................... 2.3 2.2.1 2.2.2 2.2.3 Conservation of Mass ......................................................................... 2.3 Conservation of Linear Momentum ..................................................... 2.6 Conservation of Energy ...................................................................... 2.8

2.3 2.4

HYDROSTATICS............................................................................................2.12 STEADY UNIFORM FLOW ............................................................................2.14 2.4.1 2.4.2 2.4.3 2.4.4 2.4.5 2.4.6 Introduction........................................................................................2.14 Shear Stress, Velocity Distribution, and Average Velocity .................2.16 Other Velocity Equations ...................................................................2.21 Average Boundary Shear Stress........................................................2.24 Relation Between Shear Stress and Velocity .....................................2.27 Energy and Momentum Coefficients for Rivers..................................2.28

2.5

UNSTEADY FLOW.........................................................................................2.32 2.5.1 2.5.2 2.5.3 2.5.4 Gravity Waves ...................................................................................2.32 Surges ...............................................................................................2.34 Hydraulic Jump ..................................................................................2.35 Roll Waves ........................................................................................2.35

2.6

STEADY RAPIDLY VARYING FLOW .............................................................2.38 2.6.1 2.6.2 2.6.3 2.6.4 2.6.5 Flow Through Transitions ..................................................................2.38 Specific Energy Diagram ...................................................................2.39 Specific Discharge Diagram...............................................................2.40 Transitions With Super Critical Flows ................................................2.43 Flow Over Drop Structures ................................................................2.44

2.7

FLOW IN BENDS ...........................................................................................2.45 2.7.1 2.7.2 Types of Bends..................................................................................2.45 Transverse Velocity Distribution in Bends..........................................2.46

ii

2.7.3 2.7.4 2.8

Subcritical Flow in Bends...................................................................2.47 Supercritical Flow in Bends................................................................2.49

GRADUALLY VARIED FLOW ........................................................................2.50 2.8.1 2.8.2 2.8.3 Introduction........................................................................................2.50 Classification of Flow Profiles ............................................................2.50 Standard Step Method for the Computation of Water Surface Profiles ..............................................................................................2.55

2.9

STREAM GAGING .........................................................................................2.56 2.9.1 2.9.2 2.9.3 2.9.4 Introduction........................................................................................2.56 Gaging Station...................................................................................2.56 Measuring Water Discharge ..............................................................2.60 Stage-Discharge Relation ..................................................................2.61

2.10 HYDRAULICS OF BRIDGE WATERWAYS....................................................2.62 2.10.1 2.10.2 2.10.3 2.10.4 2.10.5 Introduction........................................................................................2.62 Backwater Effects on Waterway Openings ........................................2.63 Effects of a Submerged Superstructure.............................................2.63 Effects of Supercritical Flow...............................................................2.63 Types of Flow in Bridge Openings .....................................................2.63

2.11 COMPUTER MODELS FOR HIGHWAY BRIDGES........................................2.65 2.11.1 One-Dimensional Computer Models ..................................................2.65 2.11.2 Two-Dimensional Computer Models ..................................................2.67 2.12 HYDRAULICS OF CULVERT FLOW ..............................................................2.67 2.12.1 Introduction........................................................................................2.67 2.13 ROADWAY OVERTOPPING ..........................................................................2.68 2.14 SOLVED PROBLEMS OPEN CHANNEL FLOW (SI) .....................................2.68 2.14.1 2.14.2 2.14.3 2.14.4 PROBLEM 1 Evaluation of Correction Factors α and β .....................2.68 PROBLEM 2 Velocity Profiles and Shear Stress................................2.71 PROBLEM 3 Superelevation in Bends ...............................................2.74 PROBLEM 4 Maximum Stream Constriction Without Causing Backwater..........................................................................................2.75 2.14.5 PROBLEM 5 Maximum Water Surface Elevation Upstream of a Grade Control Structure Without Backwater ......................................2.76

2.15 SOLVED PROBLEMS OPEN CHANNEL FLOW (ENGLISH) .........................2.79 2.15.1 PROBLEM 1 Evaluation of Correction Factors α and β .....................2.79 2.15.2 PROBLEM 2 Velocity Profiles and Shear Stress................................2.81 2.15.3 PROBLEM 3 Superelevation in Bends ...............................................2.84

iii

2.15.4 PROBLEM 4 Maximum Stream Constriction Without Causing Backwater..........................................................................................2.85 2.15.5 PROBLEM 5 Maximum Elevation of a Grade Control Structure Without Backwater .............................................................2.86 3. FUNDAMENTALS OF ALLUVIAL CHANNEL FLOW ................................................. 3.1 3.1 3.2 INTRODUCTION ............................................................................................. 3.1 SEDIMENT PROPERTIES AND MEASUREMENT TECHNIQUES ................. 3.1 3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.2.6 3.2.7 3.2.8 3.3 Particle Size........................................................................................ 3.1 Particle Shape .................................................................................... 3.2 Fall Velocity ........................................................................................ 3.4 Sediment Size Distribution.................................................................. 3.6 Specific Weight..................................................................................3.11 Porosity..............................................................................................3.11 Cohesion ...........................................................................................3.12 Angle of Repose ................................................................................3.12

FLOW IN SANDBED CHANNELS ..................................................................3.12 3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.3.6 3.3.7 3.3.8 3.3.9 3.3.10 Regimes of Flow in Alluvial Channels ................................................3.13 Bed Configuration ..............................................................................3.15 Plane Bed Without Sediment Movement ...........................................3.17 Ripples...............................................................................................3.17 Dunes ................................................................................................3.17 Plane Bed With Movement ................................................................3.18 Antidunes...........................................................................................3.18 Chutes and Pools ..............................................................................3.19 Regime of Flow, Bed Configuration, and Froude Number..................3.19 Bars ...................................................................................................3.19

3.4

RESISTANCE TO FLOW IN ALLUVIAL CHANNELS .....................................3.21 3.4.1 3.4.2 3.4.3 3.4.4 3.4.5 3.4.6 3.4.7 3.4.8 3.4.9 3.4.10 3.4.11 3.4.12 3.4.13 Depth.................................................................................................3.22 Slope .................................................................................................3.23 Apparent Viscosity and Density .........................................................3.23 Size of Bed Material...........................................................................3.24 Size Gradation ..................................................................................3.25 Fall Velocity .......................................................................................3.25 Shape Factor for the Reach and Cross-Section.................................3.26 Seepage Force ..................................................................................3.26 Bed Material Concentration ...............................................................3.26 Fine-Material Concentration...............................................................3.26 Bedform Predictor and Manning's n Values for Sand-Bed Streams ...3.27 How Bedform Changes Affect Highways in the River Environment....3.28 Alluvial Processes and Resistance to Flow in Coarse Material Streams .............................................................................................3.29

iv

3.5

BEGINNING OF MOTION ..............................................................................3.32 3.5.1 3.5.2 3.5.3 3.5.4 3.5.5 3.5.6 3.5.7 3.5.8 Introduction........................................................................................3.32 Theoretical Considerations ................................................................3.32 Theory of Beginning of Motion ...........................................................3.33 Shields Diagram ................................................................................3.36 Equations for Flow and Sediment Variables for Beginning of Motion .3.39 Tables for Determining Critical Velocities...........................................3.43 Figures for Determining Critical Shear Stress or Velocity ..................3.44 Summary ...........................................................................................3.45

3.6

SEDIMENT DISCHARGE MEASUREMENT...................................................3.46 3.6.1 3.6.2 3.6.3 3.6.4 3.6.5 3.6.6 3.6.7 Introduction........................................................................................3.46 Terminology.......................................................................................3.46 Suspended Sediment Discharge Measurement .................................3.47 Velocity Weighted Mean Suspended-Sediment Concentration ..........3.47 Suspended Sediment Hydrograph .....................................................3.49 Determination of Daily Suspended-Sediment Discharge....................3.49 Total Sediment Discharge..................................................................3.51

3.7

SOLVED PROBLEMS FOR ALLUVIAL CHANNEL FLOW (SI).......................3.52 3.7.1 3.7.2 3.7.3 3.7.4 PROBLEM 1 Sediment Properties and Fall Velocities .......................3.52 PROBLEM 2 Angle of Repose ...........................................................3.56 PROBLEM 3 Resistance to Flow in Alluvial Channels........................3.57 PROBLEM 4 Beginning of Motion......................................................3.59

3.8

SOLVED PROBLEMS FOR ALLUVIAL CHANNEL FLOW (ENGLISH) ..........3.61 3.8.1 3.8.2 3.8.3 3.8.4 PROBLEM 1 Sediment Properties and Fall Velocities .......................3.61 PROBLEM 2 Angle of Repose ...........................................................3.62 PROBLEM 3 Resistance to Flow in Alluvial Channels........................3.62 PROBLEM 4 Beginning of Motion......................................................3.64

4.

SEDIMENT TRANSPORT ......................................................................................... 4.1 4.1 4.2 4.3 INTRODUCTION ............................................................................................. 4.1 DEFINITIONS.................................................................................................. 4.1 GENERAL CONSIDERATIONS....................................................................... 4.3 4.3.1 4.3.2 4.3.3 4.4 4.5 Source of Sediment Transport ............................................................ 4.3 Mode of Sediment Transport .............................................................. 4.4 Total Sediment Discharge................................................................... 4.5

SUSPENDED BED SEDIMENT DISCHARGE .................................................. 4.6 BED SEDIMENT DISCHARGE........................................................................4.10 4.5.1 4.5.2 4.5.3 4.5.4 4.5.5 Meyer-Peter and Müller Bed Load Equation ......................................4.10 Einstein's Method of Computing Bed Sediment Discharge ................4.13 Comparison of Meyer-Peter and Müller and Einstein Contact Load Equations..................................................................................4.21 Colby's Method of Estimating Bed Sediment Discharge ....................4.22 Relative Influence of Variables ..........................................................4.24 v

4.6

POWER FUNCTION RELATIONSHIPS ..........................................................4.27 4.6.1 4.6.2 4.6.3 Introduction........................................................................................4.27 Basic Power Function Relationship....................................................4.27 Expanded Power Function Relationship.............................................4.29

4.7 4.8 4.9 4.10 4.11

YANG'S EQUATIONS .....................................................................................4.30 CONVERSION FACTORS...............................................................................4.31 APPLICATION OF SELECTED SEDIMENT TRANSPORT EQUATIONS .......4.32 SUMMARY ......................................................................................................4.34 SEDIMENT TRANSPORTATION ANALYSIS PROCEDURE...........................4.34 4.11.1 4.11.2 4.11.3 Step 1: Determine if Sediment Transport Computations are Necessary ...........................................................................4.34 Step 2: Determine Scope of Sediment Transport Analysis...............4.35 Step 3: Determine the Type of Analysis ...........................................4.35

4.12 SOLVED PROBLEMS FOR SEDIMENT TRANSPORT (SI) ............................4.37 4.12.1 4.12.2 4.12.3 4.12.4 4.12.5 4.12.6 4.12.7 4.12.8 Introduction.......................................................................................4.37 Problem 1 Suspended Sediment Concentration Profile ....................4.37 Problem 2 Using the Meyer-Peter and Müller Equation Calculate the Bed Sediment Discharge (Bed Load) ..........................4.38 Problem 3 Application of the Einstein Method to Calculate Total Bed-Material Discharge ...........................................................4.39 Problem 4 Calculation of Total Bed-Material Discharge Using Colby's Method .......................................................................4.39 Problem 5 Calculation of Total Bed-Material Discharge Using the Basic Power Function Relationship.............................................4.40 Problem 6 Calculation of Total Bed-Material Discharge Using the Expanded Power Function Relationship .....................................4.40 Problem 7 Calculate Total Bed-Material Discharge Using Yang's Sand Equation ............................................................4.41

4.13 SOLVED PROBLEMS FOR SEDIMENT TRANSPORT (ENGLISH) .................4.42 4.13.1 4.13.2 4.13.3 4.13.4 4.13.5 4.13.6 4.13.7 4.13.8 5. Introduction.......................................................................................4.42 Problem 1 Suspended Sediment Concentration Profile ....................4.42 Problem 2 Using the Meyer-Peter and Müller Equation Calculate the Bed Sediment Discharge (Bed Load) ..........................4.43 Problem 3 Application of the Einstein Method to Calculate Total Bed-Material Discharge ...........................................................4.44 Problem 4 Calculation of Total Bed-Material Discharge Using Colby's Method .......................................................................4.50 Problem 5 Calculation of Total Bed-Material Discharge Using the Basic Power Function Relationship.............................................4.52 Problem 6 Calculation of Total Bed-Material Discharge Using the Expanded Power Function Relationship .....................................4.53 Problem 7 Calculate Total Bed-Material Discharge Using Yang's Equation ...............................................................................4.53

RIVER MORPHOLOGY AND RIVER RESPONSE .................................................... 5.1 5.1 INTRODUCTION ............................................................................................. 5.1 vi

5.2

FLUVIAL CYCLES AND PROCESSES ........................................................... 5.1 5.2.1 5.2.2 5.2.3 5.2.4 5.2.5 Youthful, Mature, and Old Streams..................................................... 5.1 Floodplain and Delta Formations ........................................................ 5.2 Alluvial Fans ....................................................................................... 5.2 Nickpoint Migration and Headcutting .................................................. 5.3 Geomorphic Threshold ....................................................................... 5.5

5.3

VARIABILITY AND CHANGE IN ALLUVIAL RIVERS ...................................... 5.5 5.3.1 5.3.2 5.3.3 Differences Through Time .................................................................. 5.7 Differences Between Reaches............................................................ 5.8 Summary ...........................................................................................5.12

5.4

STREAM FORM AND GEOMETRY OF ALLUVIAL CHANNELS ....................5.12 5.4.1 5.4.2 5.4.3 5.4.4 5.4.5 5.4.6 5.4.7 5.4.8 Classification of River Channels ........................................................5.13 Straight River Channels .....................................................................5.17 Meandering River Channels...............................................................5.17 Braided River Channels .....................................................................5.20 River Conditions for Meandering and Braiding...................................5.21 Hydraulic Geometry of Alluvial Channels ..........................................5.23 Dominant Discharge in Alluvial Rivers ...............................................5.27 River Profiles and Bed Material..........................................................5.28

5.5

QUALITATIVE RESPONSE OF RIVER SYSTEMS ........................................5.28 5.5.1 5.5.2 5.5.3 General River Response to Change ..................................................5.29 Prediction of Channel Response to Change ......................................5.32 River Pattern Thresholds and Response ...........................................5.34

5.6

MODELING OF RIVER SYSTEMS .................................................................5.39 5.6.1 5.6.2 5.6.3 Physical Modeling ..............................................................................5.40 Computer Modeling ...........................................................................5.43 Data Needs........................................................................................5.47

5.7

HIGHWAY PROBLEMS RELATED TO GRADATION CHANGES .................5.48 5.7.1 5.7.2 5.7.3 Changes Due to Human Activities .....................................................5.48 Natural Causes ..................................................................................5.50 Resulting Problems at Highway Crossings ........................................5.51

5.8

STREAM STABILITY PROBLEMS AT HIGHWAY CROSSINGS....................5.52 5.8.1 5.8.2 5.8.3 5.8.4 5.8.5 Bank Stability.....................................................................................5.53 Stability Problems Associated With Channel Relocation....................5.54 Assessment of Stability for Relocated Streams .................................5.59 Estimation of Future Channel Stability and Behavior .........................5.60 Advances in Predicting Meander Migration ........................................5.64

vii

5.9

SOLVED PROBLEMS RIVER MORPHOLOGY AND RESPONSE .................5.67 5.9.1 5.9.2 5.9.3 5.9.4 5.9.5 5.9.6 5.9.7 5.9.8 5.9.9 PROBLEM 1 Meandering and Braiding..............................................5.67 PROBLEM 2 Classification of Alluvial Reaches .................................5.68 PROBLEM 3 Channel Response to Changes in Watershed Conditions .........................................................................................5.69 PROBLEM 4 Channel Migration Rate ................................................5.69 PROBLEM 5 At-A-Station and Downstream Hydraulic Geometry Relationships (SI) ..............................................................................5.71 PROBLEM 6 At-A-Station and Downstream Hydraulic Geometry Relationships (English) ......................................................................5.73 PROBLEM 7 Downstream Sediment Size Distribution .......................5.74 PROBLEM 8 Scale Ratios for Physical Models (SI) ...........................5.75 PROBLEM 9 Scale Ratios for Physical Models (English)...................5.76

6.

RIVER STABILIZATION AND BANK PROTECTION ................................................. 6.1 6.1 6.2 OVERVIEW ..................................................................................................... 6.1 STREAM BANK EROSION.............................................................................. 6.2 6.2.1 6.2.2 6.2.3 6.2.4 6.2.5 6.3 Causes of Streambank Failure ........................................................... 6.2 Bed and Bank Material........................................................................ 6.3 Subsurface Flow ................................................................................. 6.3 Piping of River Banks ......................................................................... 6.5 Mass Wasting ..................................................................................... 6.5

RIVER TRAINING AND STABILIZATION ........................................................ 6.6 6.3.1 6.3.2 6.3.3 6.3.4 6.3.5 Fixed Points........................................................................................ 6.6 Radius of Curvature............................................................................ 6.6 Countermeasures for Channel Instability ............................................ 6.7 Countermeasure Design Guidelines ................................................... 6.8 Protection of Training Works .............................................................. 6.9

6.4

FLOW CONTROL STRUCTURES .................................................................6.10 6.4.1 6.4.2 6.4.3 6.4.4 6.4.5 6.4.6 6.4.7 6.4.8 6.4.9 6.4.10 6.4.11 Spurs .................................................................................................6.10 Bendway Weirs..................................................................................6.12 Hardpoints .........................................................................................6.12 Retards..............................................................................................6.12 Dikes (Floodplain)..............................................................................6.14 Dikes (Channel) .................................................................................6.14 Jetties ................................................................................................6.16 Bulkheads..........................................................................................6.17 Fencing..............................................................................................6.18 Guide Banks ......................................................................................6.18 Drop Structures .................................................................................6.19

viii

...6.....2 6.............................11 SOLVED PROBLEMS FOR FILTER DESIGN (SI) ..........3 Environmental Impacts ......7 6...............2 6..................56 PROBLEM 3 Stability Factors for Riprap Design ..........................8 Bioengineering Erosion Control ......................4 6....................24 U.2 6..........6........28 Riprap Placement ..................3 6.....5.....................40 Used Tires ..............9.....5...............10..............6..............................................11........................1 6...........................6.................6...9..8..........5 Introduction............6..........................3 6........................................6.......................................6............................41 Soil Cement .7 6....1 PROBLEM 1 Filter Design .......25 Riprap Gradation and Thickness .6.........62 ix .......48 Erosion Protection in Overtoppng Flow....37 Rock-and-Wire Mattresses ..........6.6..................45 Mechanics of Overflow Erosion.......................8...6 BANK PROTECTION OTHER THAN RIPRAP .........2 6.....10.....S...........61 6.....10.................5...1 6..6........4 6.................6......6..............6........................6...............60 6...................51 6..6.................20 Stability Factor Design Methods ..................................5......................................6...............5..............39 Sacks.................................39 Articulated Concrete Block Systems (ACBs).....6......................53 Effects of Channelization ..........................................6.....8.....................55 PROBLEM 2 Riprap Design on Embankment Slopes ...............33 6.....44 6..............1 6...4 PROBLEM 1 Stability of Particles Under Downslope Flow....2 6.......55 6..................6.......9.6...6...............1 6..................................50 Overtopping Protection Systems...................36 6........6..6..........................................................6................................6......................36 Bioengineering Countermeasures....3 6.......6.................6....................21 Velocity Profile and Tractive Force ..................6..........................................6.........................6 6...................................3 6...57 PROBLEM 4 Riprap Design on an Abutment.....6................................5 6.20 6...............8 FILTERS............2 PROBLEM 2 Filter Design .........................6.61 6...6....8...............30 Riprap Failure Modes.............6 6..........6....10 SOLVED PROBLEMS FOR STABILITY OF RIPRAP (SI) ......................1 6..6........................6.42 6..54 Channel Restoration and Rehabilitation............6........6............ Army Corps of Engineers Design Equation....................................11....10..........5........6.........................................................5 6........38 Gabions .....6.......5...........55 6............8......6...............................7 Factors to Consider ..............................44 Hydraulics of Overtopping Flow ........6..................................5 RIPRAP DESIGN AND PLACEMENT..................................................6..........4 6.....................................6...........................43 OVERTOPPING FLOW ON EMBANKMENTS ..........................53 6...................................................................................................9 ENVIRONMENTAL CONSIDERATIONS ....

6.12 SOLVED PROBLEMS FOR STABILITY OF RIPRAP (ENGLISH) ..................6.64 6.12.1 6.12.2 6.12.3 6.12.4 PROBLEM 1 Stability of Particles Under Downslope Flow.................6.64 PROBLEM 2 Riprap Design on Embankment Slopes ........................6.65 PROBLEM 3 Stability Factors for Riprap Design ...............................6.67 PROBLEM 4 Riprap Design on an Abutment.....................................6.68

6.13 SOLVED PROBLEMS FOR FILTER DESIGN (ENGLISH) .............................6.70 6.13.1 PROBLEM 1 Filter Design .................................................................6.71 6.13.2 PROBLEM 2 Filter Design .................................................................6.71 7. SCOUR AT BRIDGES ............................................................................................... 7.1 7.1 INTRODUCTION ............................................................................................. 7.1 7.1.1 7.1.2 7.1.3 7.1.4 7.2 7.3 7.4 7.5 7.6 General............................................................................................... 7.1 Costs of Bridge Failure From Scour.................................................... 7.1 Bridge Scour Evaluation Program....................................................... 7.2 Comprehensive Scour Analysis .......................................................... 7.2

TOTAL SCOUR ............................................................................................... 7.4 CLEAR-WATER AND LIVE-BED SCOUR ....................................................... 7.4 ARMORING..................................................................................................... 7.6 LONG-TERM BED ELEVATION CHANGES.................................................... 7.6 GENERAL SCOUR.......................................................................................... 7.7 7.6.1 7.6.2 Contraction Scour ............................................................................... 7.8 Computer Models for General Scour .................................................7.12

7.7

LOCAL SCOUR AT PIERS.............................................................................7.12 7.7.1 7.7.2 7.7.3 7.7.4 7.7.5 7.7.6 7.7.7 7.7.8 Introduction........................................................................................7.12 Comparison of Pier Scour Equations .................................................7.13 Colorado State University's Equation.................................................7.13 FHWA HEC-18 Equation ...................................................................7.16 Pier Scour in Cohesive Bed Material..................................................7.19 Pier Scour for Other Pier Geometry, Flow Conditions, and Debris.....7.20 Topwidth of Scour Holes....................................................................7.20 Physical Model Studies......................................................................7.21

7.8

LOCAL SCOUR AT ABUTMENTS..................................................................7.21 7.8.1 7.8.2 7.8.3 7.8.4 7.8.5 7.8.6 Introduction........................................................................................7.21 Commentary on Abutment Scour Equations ......................................7.23 Abutment Site Conditions ..................................................................7.24 Abutment Shape ................................................................................7.25 Froehlich's Live-Bed Scour Equation .................................................7.25 1975 and 1990 HIRE Equation ..........................................................7.26

7.9

SCOUR PROBLEMS ......................................................................................7.27

x

8.

DATA NEEDS AND DATA SOURCES ...................................................................... 8.1 8.1 BASIC DATA NEEDS ...................................................................................... 8.1 8.1.1 8.1.2 8.1.3 8.1.4 8.1.5 8.1.6 8.1.7 8.1.8 8.1.9 8.1.10 8.2 8.3 8.4 Area Maps .......................................................................................... 8.1 Vicinity Maps....................................................................................... 8.1 Site Maps............................................................................................ 8.1 Aerial and Other Photographs ............................................................ 8.2 Field Inspection................................................................................... 8.2 Geologic Map ..................................................................................... 8.2 Climatologic Data................................................................................ 8.3 Hydraulic Data .................................................................................... 8.3 Hydrologic Data .................................................................................. 8.4 Environmental Data ............................................................................ 8.5

CHECKLIST OF DATA NEEDS ....................................................................... 8.5 DATA SOURCES ............................................................................................ 8.5 COMPUTERIZED LITERATURE AND DATA SEARCH..................................8.10 8.4.1 8.4.2 8.4.3 8.4.4 8.4.5 8.4.6 COMPENDEX....................................................................................8.10 ENVIRONMENTAL BIBLIOGRAPHY.................................................8.10 CIVIL ENGINEERING DATABASE....................................................8.11 WATER RESOURCES ABSTRACTS................................................8.11 TRIS ..................................................................................................8.11 GeoRef ..............................................................................................8.12

9.

DESIGN CONSIDERATIONS FOR HIGHWAY ENCROACHMENTS AND RIVER CROSSINGS ................................................................................................. 9.1 9.1 9.2 INTRODUCTION ............................................................................................. 9.1 PRINCIPAL FACTORS TO BE CONSIDERED IN DESIGN ............................ 9.1 9.2.1 9.2.2 9.2.3 9.2.4 9.2.5 9.2.6 9.2.7 9.2.8 9.2.9 9.2.10 9.2.11 9.2.12 9.2.13 9.3 Types of Rivers................................................................................... 9.1 Location of the Crossing or Encroachment ......................................... 9.2 River Characteristics........................................................................... 9.2 River Geometry................................................................................... 9.2 Hydrologic Data .................................................................................. 9.3 Hydraulic Data .................................................................................... 9.3 Characteristics of the Watershed........................................................ 9.3 Flow Alignment ................................................................................... 9.4 Flow on the Floodplain........................................................................ 9.4 Site Selection...................................................................................... 9.5 Channel Stability Investigations .......................................................... 9.5 Short-Term Response ....................................................................... 9.5 Long-Term Response ......................................................................... 9.6

PROCEDURE FOR EVALUATION AND DESIGN OF RIVER CROSSINGS AND ENCROACHMENTS ......................................................... 9.6 9.3.1 9.3.2 Approach to River Engineering Projects ............................................. 9.6 Project Initiation .................................................................................. 9.7

9.4

CONCEPTUAL EXAMPLES OF RIVER ENCROACHMENTS ......................... 9.8

xi

9.5

PRACTICAL EXAMPLES OF RIVER ENCROACHMENTS ............................9.20 9.5.1 9.5.2 9.5.3 9.5.4 9.5.5 9.5.6 9.5.7 9.5.8 9.5.9 9.5.10 9.5.11 9.5.12 9.5.13 9.5.14 9.5.15 9.5.16 Cimarron River, East of Okeene, Oklahoma (Example 1)..................9.20 Arkansas River, North of Bixby, Oklahoma (Example 2) ....................9.20 Washita River, North of Maysville, Oklahoma (Example 3) ................9.20 Beaver River, North of Laverne, Oklahoma (Example 4) ...................9.23 Powder River, 40 Miles East of Buffalo, Wyoming (Example 5) .........9.23 North Platte River, Near Guernsey, Wyoming (Example 6) ...............9.25 Coal Creek, Tributary of Powder River, Wyoming (Example 7)..........9.26 South Fork of Deer River at US-51 Near Halls, Tennessee (Example 8) .......................................................................................9.26 Elk Creek at SR-15 Near Jackson, Nebraska (Example 9) ................9.30 Big Elk Creek at I-90 Near Piedmont, South Dakota (Example 10)....9.32 Outlet Creek at US-101 Near Longvale, California (Example 11).......9.33 Nojoqui Creek at US-101 at Buellton, California (Example 12) ..........9.34 Turkey Creek at I-10 Near Newton, Mississippi (Example 13) ...........9.35 Gravel Mining on the Russian River, California (Example 14) ............9.36 Nowood River and Ten Sleep Creek Confluence, Wyoming (Example 15) .....................................................................................9.37 Middle Fork Powder River, Wyoming (Example 16) ...........................9.37

9.6

CONCLUDING REMARKS ON DESIGN CONSIDERATIONS .......................9.37

10. OVERVIEW EXAMPLES OF DESIGN FOR HIGHWAYS IN THE RIVER ENVIRONMENT ......................................................................................................10.1 10.1 OVERVIEW EXAMPLE 1 - BIJOU CREEK....................................................10.1 10.1.1 10.1.2 10.1.3 10.1.4 10.2 Level 1 - Reconnaissance and Geomorphic Analysis .......................10.2 Level 2 - Quantitative Engineering Analysis ......................................10.5 Riprap Design for Alternatives 1, 2, and 3........................................10.11 Design for Alternatives 4 and 5........................................................10.13

OVERVIEW EXAMPLE 2 - RILLITO RIVER ................................................10.17 10.2.1 10.2.2 10.2.3 10.2.4 10.2.5 Background .....................................................................................10.17 Level 1 - Qualitative Geomorphic Analysis ......................................10.19 Level 2 - Engineering Geomorphic Analysis ....................................10.23 Level 3 - Sediment Routing for the 100-Year Flood .........................10.31 Results of Analysis ..........................................................................10.32

10.3

BRI-STARS SEDIMENT TRANSPORT MODELING EXAMPLE ..................10.38 10.3.1 10.3.2 10.3.3 10.3.4 10.3.5 Background .....................................................................................10.38 Level 1 - Reconnaissance and Geomorphic Analysis .....................10.40 Level 2 - Hydraulic Engineering Analysis .........................................10.41 Level 2 - Scour Analysis Results......................................................10.43 Level 3 - BRI-STARS Analysis.........................................................10.45

11. REFERENCES ........................................................................................................11.1 APPENDIX A - Metric System, Conversion Factors, and Water Properties ..................... A.1 APPENDIX B - Analysis of Selected Sediment Transport Relationships........................... B.1 APPENDIX C - Index ........................................................................................................ C.1 xii

LIST OF FIGURES Figure 1.1 Figure 1.2 Figure 1.3 Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5 Figure 2.6 Figure 2.7 Figure 2.8 Figure 2.9 Geometric properties of bridge crossings .................................................... 1.3 Comparison of the 1884 and 1968 Mississippi River Channel near Commerce, Missouri.................................................................................... 1.4 Sinuosity vs. slope with constant discharge ................................................. 1.8 A river reach as a control volume ................................................................ 2.4 The control volume for conservation of linear momentum............................ 2.6 The streamtube as a control volume............................................................ 2.8 Pressure distribution in steady uniform and in steady nonuniform flow .......2.13 Pressure distribution in steady uniform flow on steep slopes ......................2.13 Steady uniform flow in a unit width channel ................................................2.14 Hydraulically smooth boundary ...................................................................2.16 Hydraulically rough boundary......................................................................2.17 Velocities in turbulent flow ..........................................................................2.17

Figure 2.10 Einstein's multiplication factor X in the logarithmic velocity equations.........2.20 Figure 2.11 Control volume for steady uniform flow.......................................................2.26 Figure 2.12 Energy and momentum coefficients for a unit width of river........................2.30 Figure 2.13 The river cross section................................................................................2.30 Figure 2.14 Definition sketch for small amplitude waves................................................2.32 Figure 2.15 Sketch of positive and negative surges.......................................................2.34 Figure 2.16 Hydraulic jump characteristics as a function of the upstream Froude number .......................................................................................................2.36 Figure 2.17 Various types of hydraulic jump ..................................................................2.36 Figure 2.18 Roll waves or slug flow ...............................................................................2.37 Figure 2.19 Transitions in open channel flow.................................................................2.38 Figure 2.20 Specific energy diagram .............................................................................2.39

xiii

Figure 2.21 Changes in water surface resulting from an increase in bed elevation........2.41 Figure 2.22 Specific discharge diagram.........................................................................2.41 Figure 2.23 Change in water surface elevation resulting from a change in width...........2.42 Figure 2.24 Flow characteristics over a drop structure...................................................2.44 Figure 2.25 Nappe profiles for supercritical flow ............................................................2.45 Figure 2.26 Schematic representation of transverse currents in a channel bed.............2.47 Figure 2.27 Definition sketch of flow around a bend ......................................................2.47 Figure 2.28 Classification of water surface profiles........................................................2.52 Figure 2.29 Examples of water surface profiles .............................................................2.54 Figure 2.30 Definition sketch for the standard step method for computation of backwater curves........................................................................................2.55 Figure 2.31 Gaging station well and shelters .................................................................2.57 Figure 2.32 Typical float recording gaging station .........................................................2.58 Figure 2.33 Typical bubble and manometer recording gaging station............................2.58 Figure 2.34 Stage vs. time hydrograph..........................................................................2.59 Figure 2.35 Definition sketch of computing area and discharge at a gaging station.......2.60 Figure 2.36 Stage-discharge relation for Schoharie Creek, New York ...........................2.61 Figure 2.37 Stage-discharge relation for a sand channel (Los Gatos Creek at El Dorado Avenue, CA) ..............................................................................2.62 Figure 2.38 Three types of backwater effect associated with bridge crossings..............2.64 Figure 2.39 Submergence of a superstructure...............................................................2.65 Figure 2.40 Types of flow encountered..........................................................................2.66 Figure 2.41 Discharge coefficient for roadway overtopping ...........................................2.69 Figure 2.42 Weir crest length determinations for roadway overtopping .........................2.69 Figure 2.43 Sketch of backwater curve over check dam................................................2.77 Figure 2.44 Sketch of backwater curve over check dam................................................2.87

xiv

Figure 3.1

Drag coefficient CD vs. particle Reynolds number Rep for spheres and natural sediments with shape factors Sp equal to 0.3, 0.5, 0.7, and 0.9 ........................................................................................................ 3.5 Nominal diameter vs. fall velocity................................................................. 3.6 Definition sketches for size-frequency characteristics of sediments............. 3.9 Angle of repose of non-cohesive materials .................................................3.13 Forms of bed roughness in sand channels .................................................3.14 Relation between water surface and bed configuration...............................3.15 Change in velocity with stream power for a sand with D50 = 0.19 mm.........3.16 Relation between regime of flow and depth for bed material with a median size equal to or less than 0.35 mm.................................................3.20 Cross-over from lower to upper flow regime based on sand size and Froude Number ...................................................................................3.20

Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9

Figure 3.10 Relation to depth to discharge for Elkhorn River near Waterloo, NE...........3.22 Figure 3.11 Apparent kinematic viscosity of water-bentonite dispersions ......................3.23 Figure 3.12 Variation of fall velocity of several sand mixtures with percent bentonite and temperature..........................................................................3.24 Figure 3.13 Relation between stream power, median fall diameter, and bed configuration and Manning's n values.......................................................3.27 Figure 3.14 Change in Manning's n with discharge for Padma River in Bangladesh....3.28 Figure 3.15 Shields Diagram: dimensionless critical shear stress.................................3.36 Figure 3.16 Shields' relation for beginning of motion .....................................................3.37 Figure 3.17 Comparison of critical shear stress as a function of grain diameter ............3.39 Figure 3.18 Critical shear stress as a function of grain diameter ...................................3.44 Figure 3.19 Critical velocity as a function of stone size..................................................3.45 Figure 3.20 Definition of sediment discharge (load) components ..................................3.46 Figure 3.21 Schematic sediment and velocity profiles ...................................................3.48 Figure 3.22a Suspended sediment sampler-D49.............................................................3.48

xv

Figure 3.22b Suspended sediment sampler-DH48 ..........................................................3.49 Figure 3.23 Gage height and suspended sediment concentration hydrograph, Colorado River near San Saba, Texas, May 1-6, 1952 ...............................3.50 Figure 3.24 Bed material size distribution curves...........................................................3.53 Figure 3.25 Size distribution curve for pebble count ......................................................3.56 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure 4.8 Figure 4.9 Classification of sediment transport in streams (rivers)................................ 4.5 Sediment transport capacity and supply curves ........................................... 4.6 Schematic sediment and velocity profiles .................................................... 4.7 Graph of suspended sediment distribution..................................................4.10 Einstein's φx vs. ψ* bed load function ..........................................................4.15 Hiding factor ...............................................................................................4.16 Einstein's multiplication factor X in the logarithmic velocity equations.........4.16 Pressure correction ....................................................................................4.17 Integral l1 in terms of E and Z .....................................................................4.18

Figure 4.10 Integral l2 in terms of E and Z .....................................................................4.19 Figure 4.11 Friction loss due to channel irregularities, as a function of sediment transport .....................................................................................................4.20 Figure 4.12 Comparison of the Meyer-Peter and Müller and Einstein methods for computing contact load..........................................................................4.21 Figure 4.13 Relation of discharge of sands to mean velocity for six median sizes of bed sands, four depths of flow, and a water temperature of 60°F...........4.22 Figure 4.14 Colby's correction curves for temperature and fine sediment......................4.23 Figure 4.15 Bed-material size effects on bed material transport ....................................4.25 Figure 4.16 Effect of slope on bed material transport ....................................................4.25 Figure 4.17 Effect of kinematic viscosity (temperature) on bed material transport .........4.26 Figure 4.18 Variation of bed material load with depth of flow.........................................4.26 Figure 4.19 Description of the average cross section ....................................................4.45

xvi

Figure 4.20 Grain size distribution of bed material.........................................................4.45 Figure 4.21 Comparison of the Einstein and Colby methods .........................................4.52 Figure 5.1 Figure 5.2 Figure 5.3 Changing slope at fan-head leading to fan-head trenching.......................... 5.4 Headcuts and nickpoints.............................................................................. 5.4 Relationship between flume slope and sinuosity (ratio between channel length and valley length) during flume experiments at constant water discharge............................................................................. 5.7 Sequence of channel changes as water discharge and sediment loads decrease ............................................................................................ 5.7 Variability of sinuosity between 1765 and 1915 for 24 Mississippi River reaches .............................................................................................. 5.8 Mississippi River reaches between Cairo, Illinois and Old River, Louisiana ..................................................................................................... 5.9 Mississippi River valley profile ..................................................................... 5.9 River Nile between Qena and Cairo showing six reaches of steep valley slope.................................................................................................5.11 River Nile valley profile ...............................................................................5.11

Figure 5.4 Figure 5.5 Figure 5.6 Figure 5.7 Figure 5.8 Figure 5.9

Figure 5.10 Plot of River Nile sinuosity and valley slope ................................................5.12 Figure 5.11 Stream properties for classification.............................................................5.14 Figure 5.12 Classification of river channels ...................................................................5.15 Figure 5.13 Channel classification showing relative stability and types of hazards encountered with each pattern....................................................................5.16 Figure 5.14 Planview and cross section of a meandering stream ..................................5.17 Figure 5.15 Definition sketch for meanders ...................................................................5.19 Figure 5.16 Empirical relations for meander characteristics...........................................5.20 Figure 5.17 Types of multi-channel streams ..................................................................5.21 Figure 5.18 Slope-discharge relationship for braiding or meandering in sandbed streams.......................................................................................................5.22 Figure 5.19 Variation of discharge at a given river cross section and at points downstream ................................................................................................5.25

xvii

Figure 5.20 Schematic variation of width, depth, and velocity with at-a-station and downstream discharge variation ..........................................................5.25 Figure 5.21 Changes in channel slope in response to an increase in sediment load at Point C....................................................................................................5.30 Figure 5.22 Changes in channel slope in response to a dam at Point C ........................5.31 Figure 5.23 Map of lower Chippewa River .....................................................................5.36 Figure 5.24 Leopold and Wolman's (1957) relation between channel patterns, channel gradient, and bankfull discharge....................................................5.37 Figure 5.25 Lane's (1957) relation between channel patterns, channel gradient, and mean discharge ..................................................................................5.37 Figure 5.26 Median bank erosion rate in relation to channel width for different types of streams .........................................................................................5.55 Figure 5.27 Encroachment on a meandering river .........................................................5.56 Figure 5.28 Erosion index in relation to sinuosity...........................................................5.61 Figure 5.29 Modes of meander loop development .........................................................5.62 Figure 5.30 Domains of meander behavior....................................................................5.65 Figure 5.31 Sinuous point bar stream ............................................................................5.67 Figure 5.32 River channels ............................................................................................5.68 Figure 5.33 Meandering river sketch .............................................................................5.70 Figure 5.34 Estimated future location of a meandering river..........................................5.70 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 6.5 Figure 6.6 Typical bank failure surfaces (a) noncohesive, (b) cohesive, (c) composite............................................................................................... 6.4 Placement of flow control structures relative to channel banks, crossing, and floodplain ..............................................................................6.11 Perspective of hard point with section detail ...............................................6.12 Retard.........................................................................................................6.13 Concrete or timber cribs .............................................................................6.13 Pile dikes ....................................................................................................6.15

xviii

Figure 6.7 Figure 6.8 Figure 6.9

Vane dike model, ground walnut shell bed, during low stage portion of test run ...................................................................................................6.15 Typical jetty-field layout ..............................................................................6.16 Steel jacks ..................................................................................................6.17

Figure 6.10 Typical guidebank.......................................................................................6.18 Figure 6.11 Definition sketch for a vertical drop.............................................................6.19 Figure 6.12 Flow and scour patterns at a sloping sill .....................................................6.20 Figure 6.13 Diagram for riprap stability conditions .........................................................6.21 Figure 6.14 Stability numbers for a 1.5 stability factor for horizontal flow along a side slope ................................................................................................6.24 Figure 6.15 Suggested gradation for riprap ...................................................................6.28 Figure 6.16 Tie-in trench to prevent riprap blanket from unraveling ...............................6.31 Figure 6.17 Rock-fill trench............................................................................................6.31 Figure 6.18 Windrow revetment, definition sketch .........................................................6.33 Figure 6.19 Riprap failure models .................................................................................6.35 Figure 6.20 Rock and wire mattress ..............................................................................6.38 Figure 6.21 Typical sand-cement bag revetment ...........................................................6.40 Figure 6.22 Articulated concrete block system ..............................................................6.41 Figure 6.23 Used tire mattress ......................................................................................6.41 Figure 6.24 Typical soil-cement bank protection............................................................6.43 Figure 6.25 Hydraulic flow zones on an embankment during overtopping flow ..............6.45 Figure 6.26 Definition sketch of variables involved in overtopping flow..........................6.46 Figure 6.27 Definition sketch for application of the momentum equation for embankment overtopping flow....................................................................6.47 Figure 6.28 Typical embankment erosion pattern with submerged flow.........................6.48 Figure 6.29 Typical embankment erosion pattern with free flow ....................................6.49

xix

Figure 6.30 Permissible velocities of various protection materials as a function of flow duration..............................................................................6.51 Figure 6.31 Full-scale test of an embankment overtopping protection system under steep-slope, high-velocity flow conditions .........................................6.52 Figure 6.32 Definition sketch for riprap on a channel bed..............................................6.56 Figure 6.33 Stability factors for various rock sizes on a side slope ................................6.59 Figure 6.34 Safety factors for various side slopes .........................................................6.59 Figure 6.35 Gradations of filter blanket for Problem 2 ...................................................6.64 Figure 6.36 Definition sketch for riprap on a channel bed..............................................6.65 Figure 6.37 Stability factors for various rock sizes on a side slope ................................6.68 Figure 6.38 Safety factors for various side slopes .........................................................6.69 Figure 6.39 Gradations of filter blanket for Problem 2 ...................................................6.74 Figure 7.1 Figure 7.2 Figure 7.3 Figure 7.4 Figure 7.5 Figure 7.6 Figure 7.7 Figure 7.8 Figure 7.9 Flow chart for scour and stream stability analysis and countermeasures......................................................................................... 7.3 Local scour depth at a pier as a function of time.......................................... 7.5 Comparison of scour formulas for variable depth ratios (y/a)......................7.14 Comparison of scour formulas with field scour measurements ...................7.14 Comparison of pier scour equations with field measurements ....................7.15 Common pier shapes..................................................................................7.17 Topwidth of scour hole................................................................................7.21 Schematic representation of abutment scour..............................................7.22 Comparison of laboratory flow characteristics to field flow conditions .........7.23

Figure 7.10 Orientation of embankment angle θ to the flow...........................................7.24 Figure 7.11 Abutment shape .........................................................................................7.25 Figure 9.1 Figure 9.2 Figure 9.3 Three-level analysis procedure for river engineering studies ....................... 9.7 Cimarron River, east of Okeene, Oklahoma (Example 1) ...........................9.21 Arkansas River, north of Bixby, Oklahoma (Example 2) .............................9.22

xx

Figure 9.4 Figure 9.5 Figure 9.6 Figure 9.7 Figure 9.8

Washita River, north of Maysville, Oklahoma (Example 3) .........................9.23 Beaver River, north of Laverne, Oklahoma (Example 4).............................9.24 Powder River, 40 miles east of Buffalo, Wyoming (Example 5) ..................9.25 North Platte River near Guernsey, Wyoming (Example 6) ..........................9.26 Coal Creek, tributary of Powder River, Wyoming (Example 7)....................9.27

Figure 9.9a Map showing South Fork of Deer River at U.S. Highway 51 Crossing (Example 8) .................................................................................9.27 Figure 9.9b Channel modifications to South Fork of Deer River at U.S. Highway 51 Crossing near Halls, Tennessee (Example 8)........................................9.28 Figure 9.9c Elevation sketch of U.S. Highway 51 Bridge (Example 8) ...........................9.29 Figure 9.10 Stage trends at Sioux City, Iowa, on the Missouri River (Example 9)..........9.31 Figure 9.11 Deflector arrangement and alignment problem on I-90 Bridge across Big Elk Creek near Piedmont, North Dakota (Example 10).........................9.32 Figure 9.12 Plan sketch of channel relocation, Outlet Creek (Example 11) ...................9.33 Figure 9.13 Plan sketch of Nojoqui Creek channel relocation (Example 12)..................9.34 Figure 9.14 Plan sketch of Turkey Creek channel relocation (Example 13) ...................9.35 Figure 9.15 Case study of sand and gravel mining (Example 14) ..................................9.36 Figure 9.16 Nowood River near Ten Sleep, Wyoming (Example 15) .............................9.38 Figure 9.17 Middle Fork Powder River at Kaycee, Wyoming (Example 16) ...................9.39 Figure 10.1 Topographic map of Bijou Creek study area ...............................................10.1 Figure 10.2 Cross sectional area versus flow depth relation ..........................................10.3 Figure 10.3 Wetted perimeter versus flow depth relation...............................................10.4 Figure 10.4 Analysis of bed material size of Bijou Creek ...............................................10.4 Figure 10.5 Gumbel's method of frequency analysis .....................................................10.6 Figure 10.6 Proposed first alternative ............................................................................10.6 Figure 10.7 Proposed second alternative ......................................................................10.7

xxi

16 Flood events at Rillito River near Tucson...10..10 The sketch of proposed riprap design...................................................28 Figure 10.24 Figure 10.40 Figure 10....................................10...................8 Figure 10.............12 Figure 10.....................14 Sketch of Craycroft Road crossing ..10........10.....29 Equal conveyance tubes of approach section.............................21 Figure 10...........................42 Figure 10.......................10.........................39 Figure 10.......16 Figure 10.....................................33 Figure 10.......................................................10..............30 Plan view of equal conveyance tubes showing velocity distribution at approach and bridge sections............................27 Figure 10......22 Bed elevation change of Rillito-Tanque Verde System (Tanque Verde flooding) .....10......................10................10........................................22 Figure 10...10.....................21 Rillito-Pantano-Tanque Verde bed sediment distribution .........18 Figure 10..................................9 Proposed fifth alternative ...........25 Alternative II for Craycroft Road Bridge crossing ....26 Alternative III for Craycroft Road Bridge crossing ....10..34 Figure 10.................................................................................10..10............................Figure 10..............................28 Cross section of proposed bridge ..20 Stage-discharge plot for Rillito River near Tucson ............27 Figure 10............................................................26 Figure 10..13 Slope-discharge relationship for Rillito River.. and IV (II shown) .............23 Bed elevation change of Rillito-Tanque Verde System (Pantano flooding)..............................................................................10....24 Alternative I for Craycroft Road Bridge crossing ....... III..10.....................................................15 Sabino Canyon Road crossing site ...10...............37 Figure 10.............11 Sketch of spur design ...........26 Figure 10.............................................................35 Figure 10..............10...10....... Arizona.....................25 Figure 10..................27 General plan view of Sabino Canyon Road crossing for alternative II.....19 100-year flood design hydrographs...................................10..............12 Rillito River system vicinity map ..........10...17 Log-normal frequency analysis for Rillito River near Tucson ..7 Figure 10.....10.............18 December 1965 flood in Rillito River and Tanque Verde Creek .........................................8 Proposed third alternative .......42 xxii ..........33 Figure 10......10....10....................

10..35 Comparison longitudinal profiles from BRI-STARS ..................47 Figure 10.......43 Figure 10.................44 Figure 10........Figure 10....................34 Three-day hydrograph for BRI-STARS analysis....................................................37 BRI-STARS model profiles for steady-state run..31 Velocity distribution at bridge crossing .............10........46 Figure 10................................46 Figure 10......................10.......10...10..........45 Figure 10.............32 Plot of total scour for example problem...36 Comparison cross-sections from BRI-STARS...................................33 Revised plot of total scour for example problem .....10.................10...................48 xxiii ......

(page intentionally left blank) xxiv .

.............................11 Detailed Computation of Superelevation in Bends .71 Velocity Profile Calculations...81 Table 2.......................1.......30 Range of Parameters Equation 4...............................1 Table 3..........................2................2........3..................................................................................3 Sediment Grade Scale......................2 Table 2............44 Sand Bed Material Size Distribution ....................1 Table 2...................................4 Table 2...52 Sand Size Bed Material Properties .......................5 Table 3..2 Table 3.........................2.........2........ 3.......6 Table 2.2..................84 Table 3..............9 Table 4........31 Maximum Permissible Velocities Proposed by Fortier and Scobey ..................................2 Table 4...........................53 Gravel Bed Material Size Distribution..............................................3.........7 Table 2...................21 Manning's Roughness Coefficients for Various Boundaries ............74 Discharge Measurement Notes..................6 Table 3...55 Summary of Applicability of Selected Sediment Transport Relations ................3 Table 3..3.....6 Typical Sediment Size Distribution for Gravel-Bed Stream ............3 Guide to Size Range for Different Types of Size Analysis.....................................30 xxv .........72 Detailed Computation of Superelevation in Bends .70 Observed Velocity Data .....2................................................................................2.........48 Developed by Simons et al.......3.....1 Table 2...............4 Table 3...........................................3....2.....3 Table 2..54 Gravel Bed Material Properties ............4...53 Discharge measurement notes .........4..............4.............................................22 Adjustment Factors for the Determination of n Values .................LIST OF TABLES Table 1............10 Velocity Profile Calculations...............43 Nonscour Velocities for Soils ..8 Table 3...............2.....1 Table 4..3...........................2......................................48 ........8 Table 2.......5 Table 2..........................27 Coefficients and Exponents of Equation 4............................................................7 Table 3.........2..............80 Observed Velocity Data ...... 3...9 Commonly Used Engineering Terms in SI and English Units.....82 Table 2.3.........25 Characteristics of Water Surface Profiles ..........................

..5.....1 Table 6. 6............4......49 Table 4..............................6....33 Chippewa River Morphology ..................72 xxvi ........................................52 Sediment Size Distribution in the St............................................... Size of Bed Sediment and Wash Load .............Table 4................ Elevation Above the Bed ........................9 Table 6.........5....6.6 Table 6........6.................4..............4..........................4 Table 6.9 Concentration vs.......................................5 Table 4..................43 Bed Material Information for Sample Problem.....4....10 Bed Material Load Calculations for Sample Problem by Applying the Colby Method (Median Diameter) .................................................50 Table 5................38 Concentration vs............7 Empirical Relations for Meanders in Alluvial Valleys .......5...............................................7 Table 5............4...........................28 Causes of Riprap Failure and Solutions .....................2 Data for Suggested Gradation ...62 Sizes of Materials.......5 Table 5............20 At-A-Station and Downstream Hydraulic Geometry Relationships ............................................2 Table 6...................2 Table 5.4.................................................32 Qualitative Response of Alluvial Channels.................6 Table 5..............8 Table 4.................................6 Table 4............................26 Derived At-A-Station and Downstream Geometry Relationships.......5..44 Hydraulic Calculations for Sample Problem by Applying the Einstein Procedure ...........6.......................................11 Bed Material Discharge Calculations for Sample Problem by Applying the Colby (Individual Size Fraction) ...............................6......7 Table 4.........8 Table 5........................................ Elevation Above the Bed .........61 Sizes of Materials...................................36 Sizes of Materials....26 Change of Variables Induced by Changes in Sediment Discharge.................................................................3 Table 5..4.........................6........5.5......71 Sizes of Materials..................................... Lawrence Seaway ................1 Table 5.......75 Factors Affecting Erosion of River Banks...............................................5................5.....42 Channel Response to Changes in Watershed and River Condition ..35 Scale Ratios for Similitude .......5...47 Table 4....................3 Table 6....46 Bed-Material Load Calculations for Sample Problem by Applying the Einstein (1950) Procedure .........................5 Table 6.4 Table 5......................................................................................

..2 Table 9......................8 Rillito River Near Tucson.............................10.................................................31 Table 10..................17 Abutment Shape Coefficients for Short Abutment Sections ....11 Table 10..............................10......................................................14 Table 10..............1 Table 8...............................1 Summary of Computer Hydraulic Conditions ............14 Table 10..10...14 Table 10........................2 Table 7.................7 Size and Filter Design for Riprap at Spur Noses and Spur Shanks.... Dominant Discharge.......36 xxvii .10 Table 10..................3 Table 7........10 Low Chord and Total Scour Requirements at Craycroft Road.. 8.............10.................3 Design of Riprap Sizes ........25 Checklist of Data Needs ...................................25 Table 10........ K2.... of the Flow .....8 Table 10..... for Pier Nose Shape.. 8....7........6 List of Data Sources .15 Table 10...........................10.......7........10..................Table 7.............4 Bank Migration for Alternative 4 ..................10 Table 10...................9 Equilibrium Slope Calculations...........4 Table 8........... K3...................... Arizona Log-Pearson Type III Frequency Analysis by USGS .......9..........................................................................2 Summary of Hydraulic Designs.................. K1................. for Bed Condition ..................7.........1 Table 7............10.......17 Correction Factor....6 Dimension of Spurs............7....... θ..................10....10........5 Summary of Spur Design... for Angle of Attack...........17 Increase in Equilibrium Pier Scour Depths...................1 Correction Factor...8 River Response to Highway Encroachments and to River Development ................................10..

(page intentionally left blank) xxviii .

and Tony Melone in the preparation of the 1975 edition. S. Johnson. Several topics have been added or expanded in this 2001 edition of the manual. Wacker.A. Specifically. D. Mahmood. Mercer and Graduate Research Assistants V. The writers wish to acknowledge the contributions made by Drs.ACKNOWLEDGMENTS This manual is a major revision of the 1990 manual. Mr. Herbert Gregory. the technical assistance of Philip Thompson. Simons. Simons. FHWA. L. Mainard A. Fiala of the Federal Highway Administration and Mainard Wacker of the Wyoming Highway Department. and M. and guidance given by the Steering Committee for the 1975 edition consisting of Frank L. L.W. and P. K. Ruff and A. Richardson. Gene R. Dean. Wyoming Highway Department. Dah-Chen Woo. J. and Lawrence J. Julien as co-authors. Simons. the writers acknowledge the contributions of Drs.B. Zevenbergen for the expanded coverage of sediment transport in Chapter 4 and Dr.V. E. D. E. the writers acknowledge the contributions of Dr. and CSU Graduate Research Assistant Noel E. and Michael E. Bormann. In addition. Lawrence J. Karaki. help. Sterling Jones.V. For the 1990 manual. Harrison.F.B. Ponce. Arneson for his critical review and valuable comments on that chapter. Copyright © Ayres Associates 2001 xxix . Li & Associates. Roger L.Y. P. Murray L.E. Larry Rundquist. Zeller. as well as the review.G. Clopper provided a new section for Chapter 6 on overtopping flow on embankments based on FHWA studies conducted in the late 1980s.M. Milo Cress. Corry.A. Harrison. Richardson. Stevens as co-authors of the 1975 manual. The writers also wish to acknowledge the help of Colorado State University Professors J.

(page intentionally left blank) xxx .

D84. D90 represent sizes for which 65.LIST OF SYMBOLS A A1 A2 An2 Ap As a a an an am ao a B. e2. e4 = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = surface area cross-sectional area for the fluid entering the control volume cross-sectional area for the fluid leaving the control volume normal cross-sectional area at the bridge crossing projected area of a pier normal to the flow area of a sediment particle acceleration or pier width or embankment length thickness of the bed layer. and 90% of the particles are smaller mean sediment size a distance x downstream from reference location mean sediment size at the reference section effective mean diameter of sediment or riprap sediment size or riprap size differential area for the fluid entering the control volume differential area for the fluid leaving the control volume thickness of riprap layer displacement of a fluid particle during time dt scour depth differential time interval energy of a system ratio of bed layer thickness to flow depth a/yo energy per unit mass moment arms of forces acting on a rock particle xxxi . similarly D65. B' bs b C C Co C1 C2 CD Cf Cr CT Cs C c c D D50 D50x D50o Dm Ds dA1 dA2 drr ds ds dt E E e e1. e3. the unmeasured zone of flow embankment length for relief bridge acceleration component normal to a streamline embankment length for main bridge amplitude of wave coefficient of width relationship dimensionless constants in Meyer-Peter. Müller equations bridge opening length exponent of width relationship sediment concentration Chezy coefficient Chezy coefficient for steady uniform flow constant for free vortex flow constant for forced vortex flow drag coefficient on a particle concentration of fine material overtopping discharge coefficient bed material sediment concentration coefficient of shear average suspended sediment concentration wave celerity coefficient of the flow depth relationship culvert diameter or sediment size sediment riprap size for which 50% by weight of the particles are smaller. 84.

I2 ib is iT J j K K1. Müller equation base backwater coefficient entrance loss coefficient for culverts correction coefficient for piers incremental correction coefficient for excentricity incremental correction coefficient for piers xxxii . Kr Kb Ke WK WKe WKp = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = eccentricity coefficient pressure force acting at section one and two functions for the transverse velocity distribution in bends buoyant force on a particle force exerted at the boundary critical Froude number for the beginning of motion of sediment drag force on a particle drag force on a rock particle fluid force on a particle gravitational force on a particle lift force on a rock particle external forces between particles Froude number shear force shear force on a particle force acting in the x direction Darcy-Weisbach friction factor Darcy-Weisbach friction factor for the grain roughness seepage force exponent of the flow depth relationship gradation coefficient acceleration due to gravity total energy entrance loss of head average head loss over a cross-section friction loss of head minimum total energy at a critical section total head velocity head headwater at a culvert horizontal side slope related to one unit vertically head loss in a hydraulic jump head loss total backwater elevation drop in water surface elevation through bridge opening integrals in the Einstein method fraction of the bed load for given equation fraction of the suspended load for a given equation fraction of total load for a given equation ratio of projected area of piers to the gross constricted area An2 exponent of the bed sediment discharge relationship coefficient of the Meyer-Peter and Müller equation coefficient for the area and volume of sediment particles coefficient in Meyer-Peter. F2(h) FB Fb Fc FD Fd Ff Fg FP Fn Fr Fs Fv Fx f f'b fs f G g H He He Hf Hmin HT Hv HW 2 Hs hL he h1* Wh I1. F2 F1(h).ec F1. K2 KB.

L3 La Lc Le Lj Ls Ls P P Pa Pb Pc M m M/N n nb no nw P p1. p2 Pb PE Pi. Po Pw pi p p Q Q QB Qb Qc Qe Qf Qn Qo Qs Qsmax QT QT q qb = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = incremental correction coefficient for skewed flow height of roughness elements coefficient of the velocity relationship total backwater coefficient for subcritical flow length of control volume length of culvert lengths for sloping sill drop structure length of abutment length of a channel along the thalweg length of embankment length of hydraulic jump upstream length of a guidebank length of the roadway crest for overspill pier length mixing length longest axis of particle intermediate axis of particle shortest axis of particle bridge opening ratio exponent of the velocity relationship ratio of lift to drag moments on a particle Manning's resistance coefficient Manning's resistance coefficient of the bed Manning's resistance coefficient for steady uniform flow Manning's resistance coefficient of the walls wetted perimeter of the flow pressure of fluid at sections one and two wetted perimeter of the bed coefficient in the Einstein's total sediment discharge method pressure of fluid inside and outside the bend wetted perimeter of the banks percentage by weight of a size fraction of sediment coefficient of the total bed sediment discharge relationship average pressure over a cross section discharge (flow rate) rate at which heat is added to the system bed load discharge in weight units water discharge quantity determining the bed load transport approach channel flow discharge approach flow discharge on the embankment formative discharge for a particular phenomena uncorrected bed sediment discharge overtopping discharge total suspended sediment discharge maximum suspended sediment discharge total bed sediment discharge total discharge unit discharge unit bed load discharge xxxiii . L2.WKs ks k k* L L L1.

Vcr Vd Vm Vmax Vn2 Vo Vp Vr Vs VT Vw V* V*c V'* V''* v1. or average velocity on a spillslope reference velocity velocity against the stone depth-average velocity at the toe of an embankment velocity of the wave shear velocity shear velocity in the channel shear velocity due to grain roughness shear velocity due to form roughness velocity of the fluid entering a control volume velocity of the fluid leaving a control volume velocity components in the s and n directions xxxiv . V2 vs.qn qs qT R R'b Re Rep R r rc ri. vn = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = uncorrected unit bed sediment discharge unit suspended sediment discharge unit total bed sediment discharge hydraulic radius hydraulic radius due to grain roughness Reynolds number Reynolds number of falling particle vector R of the direction of particle movement radius of curvature radius of curvature at the center of the stream radius of curvature for inner and outer banks in a bend coefficient of Manning's n relationship channel slope shape factor of a cross-section friction slope (also called the energy slope) friction slope to overcome grain resistance ratio of tangents of friction angle to side slope angle channel sinuosity bed slope shape factor of sediment particle shape factor of a river reach specific gravity of sediment particle bed slope at a distance x downstream of a reference location slope of water surface direction tangental to streamline wave period time temperature coefficient of the friction slope relationship internal energy associated with fluid temperature velocity vector volume critical velocity depth-average velocity at the vena contracta mean stream velocity maximum velocity average velocity at the bridge opening tangential velocity in a bend volume of a particle. ro r S Sc Sf S'f Sm Sn So Sp SR Ss Sx Sw s T T Tâ t u V V Vc. V1 v2.

vx. vy vx. vy v'x. a function of D65/w' in Einstein's method vertical distance critical depth of flow maximum depth of flow normal depth of flow total flow depth scour depth due to contraction vertical distance above bed at which the velocity is zero exponent of the Manning's n relationship exponent of the Rouse equation superelevation of flow in bends drop in water surface elevation along a channel elevation above arbitrary reference level water surface elevation at inside and outside of bend exponent of the friction slope relationship GREEK SYMBOLS α α α' β β β1 β' β βx γ γs ∆ ∆o δ δ' ε η η1 θ θ = = = = = = = = = = = = = = = = = = = = energy correction factor coefficient of the downstream decrease in slope energy correction factor for unit width momentum correction factor angle between the particle movement direction and the vertical angle of the hydraulic jump in bends momentum correction factor for a unit width coefficient of the downstream decrease in sediment size correction term in Einstein's Method specific weight of the fluid specific weight of the sediment apparent roughness of the bed dimensionless bend angle angle between the drag force and the particle movement direction thickness of the laminar sublayer in turbulent flow exponent in Laursen's equation stability number for particles on a plane bed stability number for particles on side slopes side slope angle inclination angle of a channel xxxv . zo z = = = = = = = = = = = = = = = = = = = = = = = = = = = velocity components along x and y average velocity components along x and y velocity fluctuations width of control volume or channel free surface width rate at which a fluid system does work on its surroundings rate of pressure work done by system Buoyant weight of sediment particle rate of stress work drop in water surface elevation at a drop structure lateral location in a cross-section Einstein's multiplication factor function of W/w' in Einstein's method pressure correction factor. v'y W W Wp Ws Wt WWS w X X Y y yc ymax yn yo ys y' y Z WZ Wz z zi.

3.θ1 θe κ λ λ µ ν ξ ξ ρ ρ1 ρ2 ρs σ τ τo τc φ* ψ ψ* ω ωa = = = = = = = = = = = = = = = = = = = = = = bend angle in supercritical flows angle of orientation of the embankment von Karman velocity constant wave length angle between the horizontal and the drag force vector dynamic viscosity of a fluid kinematic viscosity of a fluid correction factor. 5.31) fluid density density of a fluid entering a control volume density of a fluid leaving a control volume density of sediment particles correction coefficient for piers shear stress bed shear stress critical shear stress dimensionless sediment transport function uncorrected entrainment function entrainment function fall velocity of a sediment particle angle of the abutment xxxvi . a function of Ds/X in Einstein's method function of other parameters (Eq.

The maximum angle (as measured from the horizontal) at which gravel or sand particles can stand. Channel wholly in alluvium. ice. Individual channel of an anabranched stream. floodplain. Elongated deposits found alternately near the right and left banks of a channel. or delta. no bedrock is exposed in channel at low flow or likely to be exposed by erosion. alluvial fan. A stream whose flow is divided at normal and lower stages by large islands or. channels are more widely and distinctly separated than in a braided stream. individual islands or bars are wider than about three times water width. A fan-shaped deposit of material at the place where a stream issues from a narrow valley of high slope onto a plain or broad valley of low slope. An anabranched stream. Protective material placed on a streambed to resist scour. more rarely. The maximum flow in one year (may be daily or instantaneous). Unconsolidated material deposited by a stream in a channel. or debris rubbing against the bank. A stream which has formed its channel in cohesive or noncohesive materials that have been and can be transported by the stream. An apron designed to settle and protect the side slopes of a scour hole after settlement. alluvial stream: alluvium: alternating bars: anabranch: anabranched stream: anastomosing stream: angle of repose: annual flood: apron: apron. launching: xxxvii . An alluvial cone is made up of the finer materials suspended in flow while a debris cone is a mixture of all sizes and kinds of materials.GLOSSARY abrasion: aggradation: alluvial channel: alluvial fan: Removal of streambank material due to entrained sediment. General and progressive buildup of the longitudinal profile of a channel bed due to sediment deposition. by large bars.

Erosion-resistant materials placed directly on a streambank to protect the bank from erosion. left (right): bankfull discharge: bank protection: bank revetment: bar: base floodplain: bed: xxxviii . fills a channel to the point of overflowing. An elongated deposit of alluvium within a channel.armor (armoring): Surfacing of channel bed. The sides of a channel between which the flow is normally confined. not permanently vegetated. The low-lying lands adjacent to a stream that may become flooded due to backwater. The increase in water surface elevation relative to the elevation occurring under natural channel and floodplain conditions. Discharge that. The material used to refill a ditch or other excavation. Velocity at a given cross section determined by dividing discharge by cross sectional area. primarily placed for lower bank protection. A sudden change in the channel course that usually occurs when a stream breaks through its banks. articulated concrete mattress: average velocity: avulsion: backfill: backwater: backwater area: bank: bank. The floodplain associated with the flood with a 100-year recurrence interval. Engineering works for the purpose of protecting streambanks from erosion. on the average. usually hinged together with corrosionresistant cable fasteners. It is induced by a bridge or other structure that obstructs or constricts the free flow of water in a channel. or embankment slope to resist erosion and scour. (b) placement of a covering to resist erosion. usually associated with a flood or a catastrophic event. The side of a channel as viewed in a downstream direction.resistant layer of relatively large particles is formed on a streambed due to the removal of finer particles by streamflow. (a) Natural process whereby an erosion. or the process of doing so. The bottom of a channel bounded by banks. Rigid concrete slabs which can move without separating as scour occurs. banks.

sliding. The solid rock exposed at the surface of the earth or overlain by soils and unconsolidated material. The part of the total sediment discharge that is composed of grain sizes found in the bed and is equal to the transport capability of the flow. The area of a bridge opening available for flow. A subordinate channel of a braided stream. antidune. Bed forms are a consequence of the interaction between hydraulic forces (boundary shear stress) and the bed sediment. Material covering all or a portion of a streambank to prevent erosion. as measured below a specified stage and normal to the principal direction of flow. dune. or skipping along the bed or very close to it. A stream whose flow is divided at normal stage by small mid-channel bars or small islands. A rock fragment whose diameter is greater than 250 mm. such as a ripple. The quantity of bed load passing a cross section of a stream in a unit of time. bed layer: bed load: bed load discharge (or bed load): bed material: bedrock: bed sediment discharge: bed shear (tractive force): bed slope: blanket: boulder: braid: braided stream: bridge opening: bridge waterway: xxxix . several grain diameters thick (usually two) immediately above the bed.bed form: A recognizable relief feature on the bed of a channel. Sediment that is transported in a stream by rolling. the individual width of bars and islands is less than about three times water width. The inclination of the channel bottom. plane bed. The force per unit area exerted by a fluid flowing past a stationary boundary. The cross-sectional area beneath a bridge that is available for conveyance of water. A flow layer. Material found in and on the bed of a stream (May be transported as bed load or in suspension). or bar. a braided stream has the aspect of a single large channel within which are subordinate channels. considered to be within the bed layer (contact load).

braiding. with particular reference to the degree of sinuosity. and anabranching. Behavior of a channel with respect to shifting. A low dam or weir across a channel used to control stage or degradation. A vertical. The aspect of a stream channel in plan view. See drainage basin. Excessive constriction of flow which may cause severe backwater effect. or near vertical. The removal of flows by natural or artificial means from a natural length of channel. also may serve to protect against erosion.004 mm. flow-control measures.bulk density: bulkhead: bulking: catchment: causeway: caving: cellular-block mattress: Density of the water sediment mixture (mass per unit volume). channel: channelization: channel diversion: channel pattern: channel process: check dam: choking (of flow): clay (mineral): clay plug: xl . including both water and sediment. Interconnected concrete blocks with regular cavities placed directly on a streambank or filter to resist erosion. The collapse of a bank caused by undermining due to the action of flowing water. wall that supports a bank or an embankment. Straightening or deepening of a natural channel by artificial cutoffs. A cutoff meander bend filled with fine grained cohesive sediments. grading. A particle whose diameter is in the range of 0. The bed and banks that confine the surface flow of a stream. or diversion of flow into an engineered channel. Increasing the water discharge to account for high concentrations of sediment in the flow. Rock or earth embankment carrying a roadway across water.00024 to 0. The cavities can permit bank drainage and the growth of vegetation where synthetic filter fabric is not used between the bank and mattress. erosion and sedimentation.

The contraction can be caused by the bridge or from a natural narrowing of the stream channel. The inertial force caused by the Earth's rotation that deflects a moving body to the right in the Northern Hemisphere.clear-water scour: Scour at a pier or abutment (or contraction scour) when there is no movement of the bed material upstream of the bridge crossing at the flow causing bridge scour. A fragment of rock whose diameter is in the range of 64 to 250 mm. channel reach or dam. such as a bridge crossing. delay or reduce the severity of hydraulic problems. with limited flow capacity in which the upstream water surface elevation is related to discharge. An instrument used to measure flow velocity. The junction of two or more streams. A frame structure filled with earth or stone ballast. The minimum amount of shear stress required to initiate soil particle motion. designed to reduce energy and to deflect streamflow away from a bank or embankment. cobble: concrete revetment: confluence: constriction: contact load: contraction: contraction scour: Coriolis force: countermeasure: crib: critical shear stress: crossing: cross section: current: current meter: xli . A section normal to the trend of a channel or flow. A natural or artificial control section. Water flowing through a channel. Contraction scour. This component of scour results from a contraction of the flow area at the bridge which causes an increase in velocity and shear stress on the bed at the bridge. in a natural channel or at a bridge crossing. also crossover or riffle. The relatively short and shallow reach of a stream between bends. involves the removal of material from the bed and banks across all or most of the channel width. Sediment particles that roll or slide along in almost continuous contact with the streambed (bed load). A measure intended to prevent. The effect of channel or bridge constriction on flow streamlines. Unreinforced or reinforced concrete slabs placed on the channel bed or banks to protect it from erosion.

spur. either natural or artificial. The discharge that is selected as the basis for the design or evaluation of a hydraulic structure. over a relatively long channel length. vegetation. An impermeable linear structure for the control or containment of overbank flow. or (b) deflect erosive current away from the streambank (impermeable dike). The dominant formative discharge depends on the maximum and mean discharge. usually of sheet piling or concrete.5 years in many natural channels. For hydraulic geometry relationships. and it may be surrounded by water during floods. A wall. A general and progressive (long-term) lowering of the channel bed due to erosion. and bed. Volume of water passing through a channel during a given time. duration of flow. jetty): discharge: dominant discharge: xlii . (b) That discharge which determines the principal dimensions and characteristics of a natural channel. that extends down to scour-resistant material or below the expected scour depth. (b) A natural or artificial channel which develops across the neck of a meander loop (neck cutoff) or across a point bar (chute cutoff). (a) The discharge of water which is of sufficient magnitude and frequency to have a dominating effect in determining the characteristics and size of the stream course. The vertical distance a streambed is lowered by scour below a reference elevation. it is taken to be the bankfull discharge which has a return period of approximately 1. (a) A direct channel. A dike-trending parallel with a streambank differs from a levee in that it extends for a much shorter distance along the bank. thus encouraging sediment deposition along the bank (permeable dike). connecting two points on a stream. transported by a stream. thereby shortening the original length of the channel and increasing its slope.cut bank: cutoff: The concave wall of a meandering stream. and flood frequency. or trash. Discharge averaged over one day (24 hours). such as logs. Floating or submerged material. A structure extending from a bank into a channel that is designed to: (a) reduce the stream velocity as the current passes through the dike. channel. cutoff wall: daily discharge: debris: degradation (bed): depth of scour: design flow (design flood): dike: dike (groin.

bank for the purpose of resisting erosion or providing temporary stabilization until vegetation is established. sometimes in double rows. entrenched stream: ephemeral stream: equilibrium scour: erosion: erosion control matting: fabric mattress: fall velocity: fascine: fetch: fetch length: fill slope: xliii .) placed or sprayed on a stream. a streambank protection technique consisting of wire mesh or timber attached to a series of posts. Side or end slope of an earth-fill embankment. or other materials. Where a fill-slope forms the streamward face of a spill-through abutment. etc. As used here. Grout-filled mattress used for streambank protection. sometimes used synonymously with fetch length. A stream or reach of stream that does not flow for parts of the year. jute.drainage basin: drift: eddy current: An area confined by drainage divides. watershed). the term includes intermittent streams with flow less than perennial. The velocity at which a sediment particle falls through a column of still water. Displacement of soil particles due to water or wind action.. A matrix of willow or other natural material woven in bundles and used as a filter. the space between the rows may be filled with rock. The horizontal distance (in the direction of the wind) over which wind generates waves and wind setup. such as the circular water movement that occurs when the main flow becomes separated from the bank. it is regarded as part of the abutment.g. Also. Fibrous matting (e. The area in which waves are generated by wind having a rather constant direction and speed. Stream cut into bedrock or consolidated deposits. paper. Alternative term for vegetative "debris. often having only one outlet for discharge (catchment." A vortex-type motion of a fluid flowing contrary to the main current. brush. Scour depth in sand-bed stream with dune bed about which live bed pier scour level fluctuates due to variability in bed material transport in the approach flow.

Flow characteristics (discharge. gravel. but some are perennial. Typically associated with mountain streams or highly disturbed urbanized catchments. A graph indicating the probability that the annual flood discharge will exceed a given magnitude. clays and sand could be considered wash load in coarse gravel and cobble-bed channels. or graded rock) placed between bank revetment (or bed protection) and soil for the following purposes: (1) to prevent the soil from moving through the revetment by piping. depth. or the recurrence interval corresponding to a given magnitude. the fine-sediment load is finer than 0. That part of the total sediment load that is composed of particle sizes finer than those represented in the bed (wash load). A structure either within or outside a channel that acts as a countermeasure by controlling the direction. or velocity of flowing water. (2) to prevent the revetment from sinking into the soil. A nearly flat. or duration) that are associated with a hydraulic problem or that can reasonably be considered of sufficient magnitude to cause a hydraulic problem or to test the effectiveness of a countermeasure. or erosion. Geosynthetic fabric that serves the same purpose as a granular filter blanket. Most flashy streams are ephemeral. as indicated by a sharply peaked hydrograph. velocity. Silts. and (3) to permit natural seepage from the streambank. filter blanket: filter fabric (cloth): fine sediment load: flanking: flashy stream: flood-frequency curve: floodplain: flow-control structure: flow hazard: flow slide: xliv . Stream characterized by rapidly rising and falling stages. stage. alluvial lowland bordering a stream.filter: Layer of fabric (geotextile) or granular material (sand.062 mm for sand-bed channels. extrusion. A layer of graded sand and gravel laid between fine-grained material and riprap to serve as a filter. that is subject to frequent inundation by floods. Normally. A flow slide on a channel bank can result in a bank failure. thus preventing the buildup of excessive hydrostatic pressure. Erosion around the landward end of a stream stabilization countermeasure. Saturated soil materials which behave more like a liquid than a solid.

drift. and the changes that take place due to erosion and deposition. A geomorphic term used for streams that have apparently achieved a state of equilibrium between the rate of sediment transport and the rate of sediment supply throughout long reaches. General scour is a lowering of the streambed across the stream or waterway at the bridge. check dam): graded stream: gravel: xlv . or the sediment deposition zone. and (3) alluvial fans. (2) tributary and mainstem river channels or sediment transfer zone. the gabion becomes a flexible and permeable unit with which flow. A basket or compartmented rectangular container made of wire mesh. The vertical distance above a design stage that is allowed for waves.fluvial geomorphology: fluvial system: The science dealing with the morphology (form) and dynamics of streams and rivers. That is. When filled with cobbles or other rock of suitable size. and other contingencies. valley fills and deltas. Structure placed bank to bank across a stream channel usually with its central axis perpendicular to flow) for the purpose of controlling bed slope and preventing scour or headcutting. The natural river system consisting of (1) the drainage basin. A dimensionless number that represents the ratio of inertial to gravitational forces in open channel flow. surges. A rock fragment whose diameter ranges from 2 to 64 mm. That science that deals with the form of the Earth. watershed. or sediment source area. the general configuration of its surface. freeboard: Froude Number: gabion: general scour: geomorphology/ morphology: grade-control structure (sill. the depth of scour may be deeper in some parts of the cross section.and erosion-control structures can be built. This lowering may be uniform across the bed or non-uniform. General scour may result from contraction of the flow or other general scour conditions such as flow around a bend.

The cross-sectional area of a stream divided by its wetted perimeter. accommodate." "jetty." etc. weirs. tidal flow. Channel degradation associated with abrupt changes in the bed elevation (headcut) that generally migrates in an upstream direction. grout: guide bank: hardpoint: headcutting: helical flow: hydraulics: hydraulic model: hydraulic problem: hydraulic radius: hydraulic structures: xlvi . A fluid mixture of cement and water or of cement. and the ground. The facilities used to impound. Many names are given to this structure. the most common being "spur. and water used to fill joints and voids. or wave action such that the integrity of the highway facility is destroyed. structures. such as dams. their net effect is to remove soil particles from the cut bank and deposit this material on a point bar." "spur dike." "transverse dike. or impermeable. damaged. A small-scale physical or mathematical representation of a flow situation. or endangered. intakes. An effect of streamflow. A dike extending upstream from the approach embankment at either or both sides of the bridge opening to direct the flow through the opening. The applied science concerned with the behavior and flow of liquids. These secondary-type currents are of most significance as flow passes through a bend. channels.groin: A structure built from the bank of a stream in a direction transverse to the current to redirect the flow or reduce flow velocity. Some guidebanks extend downstream from the bridge (also spur dike). convey or control the flow of water. channels. and bridges. culverts. semi-permeable. A streambank protection structure whereby "soft" or erodible materials are removed from a bank and replaced by stone or compacted clay. Some hard points protrude a short distance into the channel to direct erosive currents away from the bank. Hard points also occur naturally along streambanks as passing currents remove erodible materials leaving nonerodible materials exposed. especially in pipes. sand. Three-dimensional movement of water particles along a spiral path in the general direction of flow. Groins may be permeable.

some rows generally parallel with the banks and some perpendicular thereto or at an angle.) downslope or into a scoured area. lake. A permanently vegetated area. rock. slag. etc. having an overlapping or shingled pattern. A stream which has deepened its channel through the bed of the valley floor. or to protect against erosion. or to protect a harbor (also spur). Release of undercut material (stone riprap. Jack fields may be placed outside or within a channel. A stretch of stream with an incised channel that only rarely overflows its banks. or dams. that divides the flow of a stream. so placed as to induce bank building. The science concerned with the occurrence. (a) An obstruction built of piles. Masses or sheets of ice formed on the frozen surface of a river or floodplain. water under hydrostatic pressure is forced to the surface where it freezes. emergent at normal stage. In reference to stream bed sediment particles. A device for flow control and protection of banks against lateral erosion consisting of three mutually perpendicular arms rigidly fixed at the center. Erosion in which the removal of material is extended horizontally as contrasted with degradation and scour in a vertical direction. When shoals in the river are frozen to the bottom or otherwise dammed. by channel avulsion. distribution. culverts. Islands originate by establishment of vegetation on a bar. (b) A similar obstruction to influence stream. The lowest point in the channel cross section or at flow control devices such as weirs. or at the junction of minor tributary with a larger stream. or other material extending from a bank into a stream. incised reach: incised stream: invert: island: jack: jack field: jetty: lateral erosion: launching: xlvii . and concrete jacks are made of reinforced concrete beams.hydrograph: hydrology: imbricated: icing: The graph of stage or discharge against time. and circulation of water on the earth. rubble. so that the floodplain is a terrace. Kellner jacks are made of steel struts strung with wire. Rows of jacks tied together with cables. or tidal currents.

that confines flow during high-water periods. one flowing clockwise and the other counter-clockwise. Amount of sediment being moved by a stream. elevations of the water surface or the thalweg are plotted against distance as measured from the mouth or from an arbitrary initial point.levee: An embankment. A blanket or revetment of materials interwoven or otherwise lashed together and placed to cover an area subject to scour. A meander in a river consists of two consecutive loops. An individual loop of a meandering or sinuous stream lying between inflection points with adjoining loops. The profile of a stream or channel drawn along the length of its centerline. measured between center lines of channels. That portion of a streambank having an elevation less than the mean water level of the stream. The distance between lines drawn tangent to the extreme limits of successive fully developed meanders. The distance between points of maximum curvature of successive meanders of opposite phase in a direction normal to the general course of the meander belt. spurs. A numerical representation of a flow situation using mathematical equations (also computer model). In drawing the profile. Removal of material from around piers. abutments. The distance along a stream between corresponding points of successive meanders. The radius of a circle inscribed on the centerline of a meander loop. and embankments caused by an acceleration of flow and resulting vortices induced by obstructions to the flow. Scour at a pier or abutment (or contraction scour) when the bed material in the channel upstream of the bridge is moving at the flow causing bridge scour. The ratio of meander width to meander length. thus preventing overflow into lowlands. generally landward of top bank. live-bed scour: load (or sediment load): local scour: longitudinal profile: lower bank: mathematical model: mattress: meander or full meander: meander amplitude: meander belt: meander length: meander loop: meander ratio: meander radius of curvature: xlviii .

compacted asphalt. The amplitude of a fully developed meander measured from midstream to midstream. and soil-cement. concentric ridges and swales on a floodplain. A stream having a sinuosity greater than some arbitrary value. The term also implies a moderate degree of pattern symmetry. Change in position of a channel by lateral erosion of one bank and simultaneous accretion of the opposite bank. The abandoned former meander loop that remains after a stream cuts a new. The water stage prevailing during the greater part of the year. The particle diameter of the 50th percentile point on a size distribution curve such that half of the particles (by weight. shorter channel across the narrow neck of a meander. or volume) are larger and half are smaller (D50. Common pavements used on streambanks are concrete. Equivalent spherical diameter of a hypothetical sphere of the same volume as a given sediment particle. marking the successive positions of former meander loops. usually impermeable. A soft. Streambank surface covering. Water movement that overtops the bank either due to stream stage or to overland surface water runoff. Often bow-shaped or horseshoe-shaped. imparted by regularity of size and repetition of meander loops. The portion of a streambank having an elevation approximately the same as that of the mean water level of the stream. The channel generally exhibits a characteristic process of bank erosion and point bar deposition associated with systematically shifting meanders. designed to serve as protection against erosion. saturated mixture mainly of silt and clay. number.) A bar lacking permanent vegetal cover that divides the flow in a channel at normal stage.meander scrolls: meander width: meandering stream: Low. A low ridge that slopes gently away from the channel banks that is formed along streambanks during floods by deposition. median diameter: mid-channel bar: middle bank: migration: mud: natural levee: nominal diameter: nonalluvial channel: normal stage: overbank flow: oxbow: pavement: xlix . A channel whose boundary is in bedrock or non-erodible material.

An alluvial deposit of sand or gravel lacking permanent vegetal cover occurring in a channel at the inside of a meander loop. A type of countermeasure composed of rock-filled wire fabric supported by steel rails or posts driven into streambed. perennial stream: phreatic line: pile: pile dike: piping: point bar: poised stream: probable maximum flood: quarry-run stone: railbank protection: rapid drawdown: reach: l . but for practical engineering purposes. Removal of soil material through subsurface flow of seepage water that develops channels or "pipes" within the soil bank. A very rare flood discharge value computed by hydrometeorological methods. concrete. Stone as received from a quarry without regard to gradation requirements. usually somewhat downstream from the apex of the loop. An elongated member. that serves as a structural component of a river-training structure. the stream may be considered stable.paving: peaked stone dike: Covering of stones on a channel bed or bank (used with reference to natural covering). and channel dimensions without any noticeable raising or lowering of its bed (stable stream). usually made of timber. or steel. The upper boundary of the seepage water surface landward of a streambank. maintains its slope. A segment of stream length that is arbitrarily bounded for purposes of study. Lowering the water against a bank more quickly than the bank can drain without becoming unstable. as a whole. usually in connection with major hydraulic structures. A type of permeable structure for the protection of banks against caving. consists of a cluster of piles driven into the stream. braced and lashed together. Riprap placed parallel to the toe of a streambank (at the natural angle of repose of the stone) to prevent erosion of the toe and induce sediment deposition behind the dike. A stream which. depths. A stream or reach of a stream that flows continuously for all or most of the year. Such condition may be temporary from a geological point of view.

An opening in an embankment on a floodplain to permit passage of overbank flow. sediment regime. induce deposition. zone. a state of equilibrium with respect to erosion and deposition. exceedance interval). etc. A streambank protection method consisting of a continuous stone toe-fill along the base of a bank slope with intermittent fillets of stone placed perpendicular to the toe and extending back into the natural bank. vegetation. tidal regime. Typically. (used also to mean a set of physical characteristics of a river). (See bank revetment).recurrence interval: regime: The reciprocal of the annual probability of exceedance of a hydrologic event (also return period. The condition of a stream or its channel with regard to stability. or slope. more or less. shallow flow area extending across a streambed in which the surface of flowing water is broken by waves or ripples. riffles alternate with pools along the length of a stream channel. intended to reduce flow velocity. as in flow regime. A change in channel characteristics resulting from such things as changes in imposed flows. regime change: regime channel: regime formula: reinforced-earth bulkhead: reinforced revetment: relief bridge: retard (retarder structure): revetment: riffle: riparian: li . Also. A natural. or deflect flow from the bank. Rigid or flexible armor placed to inhibit scour and lateral erosion. Alluvial channel that has attained. the general pattern of variation around a mean condition. sediment loads.). A permeable or impermeable linear structure in a channel parallel with the bank and usually at the toe of the bank. etc. channel regime. A stream is in regime if its channel has reached an equilibrium form as a result of its flow characteristics. A formula relating stable alluvial channel dimensions or slope to discharge and sediment characteristics. A retaining structure consisting of vertical panels and attached to reinforcing elements embedded in compacted backfill for supporting a streambank. Pertaining to anything connected with or adjacent to the banks of a stream (corridor.

any structure configuration constructed in a stream or placed on. induce scour.g. Sediment bounced along the streambed by energy and turbulence of flow. Riprap has also been applied to almost all kinds of armor. Numerical measure of the frictional resistance to flow in a channel. Engineering works with or without the construction of embankment. grouted riprap. built along a stream or reach of stream to direct or to lead the flow into a prescribed channel. also the rock or broken concrete suitable for such use. or in the vicinity of a streambank that is intended to deflect currents. sand. concrete. river training: rock-and-wire mattress: roughness coefficient: rubble: runoff: sack revetment: saltation load: sand: scour: sediment or fluvial sediment: sediment concentration: sediment discharge: lii . or nylon) filled with mortar. adjacent to. sacked concrete. Rough. The fragments may consist of broken concrete slabs. irregular fragments of materials of random size used to retard erosion. induce sediment deposition. including wire-enclosed riprap. and by other moving particles. Sacks (e. stone or other available material used as protection against erosion. Erosion of streambed or bank material due to flowing water. Also.062 to 2. Discharge may be limited to certain sizes of sediment or to a specific part of the cross section. That part of precipitation which appears in surface streams of either perennial or intermittent form. suspended. and concrete slabs. often considered as being localized (see local scour. masonry. A rock fragment whose diameter is in the range of 0. The quantity of sediment that is carried past any cross section of a stream in a unit of time. Fragmental material transported. paper. A flat wire cage or basket filled with stone or other suitable material and placed as protection against erosion. contraction scour.riprap: Layer or facing of rock or broken concrete which is dumped or placed to protect a structure or embankment from erosion. as in the Manning's or Chezy's formulas. total scour).. or in some other way alter the flow and sediment regimes of the stream. or deposited by water. Weight or volume of sediment relative to the quantity of transporting (or suspending) fluid.0 mm. or other suitable refuse. burlap.

seepage: shear stress: shoal: sill: silt: sinuosity: slope (of channel or stream): slope protection: sloughing: slope-area method: slump: soil-cement: sorting: liii . but usually occurs when a bank or an underlying stratum is saturated. Sliding or collapse of overlying material. (b) A low structure built across an effluent stream. A sudden slip or collapse of a bank.sediment load: sediment yield: Amount of sediment being moved by a stream. paving. A particle whose diameter is in the range of 0.004 to 0. The slow movement of water through small cracks and pores of the bank material. same ultimate effect as caving. A designed mixture of soil and Portland cement compacted at a proper water content to form a blanket or structure that can resist erosion. diversion channel or outlet to reduce flow or prevent flow until the main stream stage reaches the crest of the structure. This outflow is equal to the sediment discharge from the drainage area. vegetation. Fall per unit length along the channel centerline or thalweg. Any measure such as riprap.062 mm. generally in the vertical direction and confined to a short distance. The total sediment outflow from a watershed or a drainage area at a point of reference and in a specified time period. slipping or caving. See unit shear force. brush or other material intended to protect a slope from erosion. A relatively shallow submerged bank or bar in a body of water. (a) A structure built under water. or to withstand external hydraulic pressure. probably due to the substratum being washed out or having become unable to bear the weight above it. revetment. The ratio between the thalweg length and the valley length of a stream. Progressive reduction of size (or weight) of particles of the sediment load carried down a stream. across the deep pools of a stream with the aim of changing the depth of the stream. A method of estimating unmeasured flood discharges in a uniform channel reach using observed high-water levels.

Any technique used to prevent erosion or failure of a streambank. Natural cobbles. that is. A condition of a channel when. A permeable or impermeable linear structure that projects into a channel from the bank to alter flow direction. chemical reactions. spread footing: spur: spur dike: stability: stable channel: stage: stone riprap: stream: streambank erosion: streambank failure: streambank protection: suspended sediment discharge: liv . Sudden collapse of a bank due to an unstable condition such as removal of material at the toe of the bank by scour. Removal of soil particles or a mass of particles from a bank surface due primarily to water action. See guide bank. induce deposition. ice and debris abrasion. the term is sometimes applied to a natural channel or drainage course formed by flowing water whether it is occupied by water or not. A body of water that may range in size from a large river to a small rill flowing in a channel. boulders.spill-through abutment: A bridge abutment having a fill slope on the streamward side. Other factors such as weathering. degradation. though it may change slightly at different times of the year as the result of varying conditions of flow and sediment charge. or bank erosion (a graded stream). there is no appreciable change from year to year. or reduce flow velocity along the bank. The term originally referred to the "spill-through" of fill at an open abutment but is now applied to any abutment having such a slope. accretion balances erosion over the years. A condition that exists when a stream has a bed slope and cross section which allows its channel to transport the water and sediment delivered from the upstream watershed without aggradation. Water-surface elevation of a stream with respect to a reference elevation. The quantity of sediment passing through a stream cross section above the bed layer in a unit of time suspended by the turbulence of flow (suspended load). By extension. or rock dumped or placed as protection against erosion. A pier or abutment footing that transfers load directly to the earth. and land use changes may also directly or indirectly lead to bank erosion.

logs. respectively. Loose stones laid or dumped at the toe of an embankment. Stone. The completed mattress is then placed on the bank of a stream and weighted with ballast.. etc. The sum of long-term degradation. The sum of suspended load and bed load or the sum of bed material load and wash load of a stream (total load). with each leg making an angle of 109. general (contraction) scour. or masonry material placed in a trench dug behind and parallel to an eroding streambank. When the erosive action of the stream reaches the trench. poles. The drag or shear on a streambed or bank caused by passing water which tends to move soil particles along with the streamflow. or masonry or concrete wall built at the junction of the bank and the bed in channels or at extremities of hydraulic structures to counteract erosion. Bank protection component of precast concrete consisting of four legs joined at a central joint. A revetment made of brush. substrate. and local scour. Structure placed between revetment and bank to prevent flanking. That portion of a stream cross section where the lower bank terminates and the channel bottom or the opposite lower bank begins. supercritical flow: tetrahedron: tetrapod: Material underlying that portion of the streambed which is subject to direct action of the flow.sub-bed material: subcritical.with the other three. or lumber interwoven or otherwise lashed together. thalweg: tieback: timber or brush mattress: toe of bank: toe protection: total scour: total sediment load: tractive force: trench-fill revetment: lv . Component of river-training works made of six steel or concrete struts fabricated in the shape of a pyramid.5. concrete. Open channel flow conditions with Froude Number less than and greater than unity. The line extending down a channel that follows the lowest elevation of the bed. Also. the material placed in the trench armors the bank and thus retards further erosion. groin.

Pa (N/m ) or 2 (lb/ft ). Usually in units of stress. The time rate of flow usually expressed in m/s (ft/sec). with no definite meanders or braided pattern. this force is equal to a component of the gravity force acting in a direction parallel to the channel bed 2 on a unit wetted area. Flow of constant cross section and velocity through a reach of channel at a given time.g. erosion and deposition. For uniform flow. The maximum depth of scour attained for a given flow condition. An abutment. May require multiple flow events and in cemented or cohesive soils may be achieved over a long time period. Both the energy slope and the water slope are equal to the bed slope under conditions of uniform flow. usually with wingwalls. The average velocity is the velocity at a given cross section determined by dividing discharge by cross-sectional area. Flow of variable discharge and velocity through a cross section with respect to time.turbulence: Motion of fluids in which local velocities and pressures fluctuate irregularly in a random manner as opposed to laminar flow where all particles of the fluid move in distinct and separate lines. The portion of a streambank having an elevation greater than the average water level of the stream.. or local at a point). Turbulent eddy in the flow generally caused by an obstruction such as a bridge pier or abutment (e. that has no fill slope on its streamward side. A thalweg whose position in the channel shifts during floods and typically serves as an inset channel that conveys all or most of the stream flow at normal or lower stages. horseshoe vortex). A channel exhibiting a more or less non-systematic process of channel shifting. The force or drag developed at the channel bed by flowing water. ultimate scour: uniform flow: unit discharge: unit shear force (shear stress): unsteady flow: upper bank: velocity: vertical abutment: vortex: wandering channel: wandering thalweg: lvi . Discharge per unit width (may be average over a cross section.

See drainage basin. measured normal to the principal direction of flow. Width (area) of bridge opening at (below) a specified stage. openings are of suitable size and shape to enclose rock or broken concrete or to function on fence-like spurs and retards. As the windrow is undercut. Wire woven to form a mesh.wash load: Suspended material of very small size (generally clays and colloids) originating primarily from erosion on the land slopes of the drainage area and present to a negligible degree in the bed itself. watershed: waterway opening width (area): weephole: windrow revetment: wire mesh: lvii . where used as an integral part of a countermeasure. thus armoring the bank. A row of stone placed landward of the top of an eroding streambank. A hole in an impermeable wall or revetment to relieve the neutral stress or pore pressure in the soil. the stone is launched downslope.

(page intentionally left blank) lviii .

Encroachment is any occupancy of the river and floodplain for highway use. especially in populated areas. sediment transport.Experience.CHAPTER 1 INTRODUCTION The purpose of this chapter is to lay the groundwork for application of the concepts of open-channel flow.1 CLASSIFICATION OF RIVER CROSSINGS AND ENCROACHMENTS The objective in this document is to consider the fluvial. geomorphic. maintenance. Some bridges and culverts must accommodate the passage of livestock and farm equipment during periods of low flow. All of these factors. Basic definitions of terms and notations adopted for use in this document have been presented in the preceding section (Glossary) for rapid reference. and many more. and river mechanics to the design. longitudinal encroachments. Encroachments usually present no problems during normal stages. and (3) Finite Element Surface Water Modeling System (FESWMS). Additionally. Still other bridges require short spans with long approaches and numerous piers for economic reasons. the possible flow conditions. "Design of Riprap Revetment" (Brown and Clyde 1989). Related NHI courses include: (1) River Engineering for Highway Encroachments. 2001). 1. and the various geometric conditions aids the engineer in selecting the design criteria for the conditions encountered. fluvial geomorphology. sediment transport. the possible channel shapes. contribute to the difficulty in generalizing the design for all highway encroachments. A classification of encroachments based on prominent features is helpful. Flood protection requirements vary from site to site. Selection and Design Guidance" (Lagasse et al. This manual is a basic reference for related Federal Highway Administration (FHWA) hydraulic publications and National Highway Institute (NHI) Hydraulics Courses. these important terms and variables are defined and explained as they are encountered. Other bridges require low embankments for aesthetic appeal. "Stream Stability at Highway Structures" (Lagasse et al. "Evaluating Scour at Bridges" (Richardson and Davis 2001). "Bridge Scour and Stream Instability Countermeasures . Classifying the regions requiring protection. stabilization works and road approaches. including bridge locations. alignments.1 . hydraulic. 2001). and environmental aspects of highway encroachments. the possible types of protection. and environmental problems associated with highway crossings and encroachments. but require special protection against floods. 1. Some of these publications are: "Hydraulics of Bridge Waterways" (Bradley 1978). (2) Stream Stability and Scour at Highway Bridges.

In many locations the highway must encroach on the main channel itself and the channel is partly filled to allow room for the roadway. scour. dimensions. and flow conditions. channel shape. constricting the low-flow channel. even when the extra cost of flood protection is included. abutment types.1. unchanging in shape. abutment setback. highways generally must impose a degree of encroachment. piers. Often. dual bridges for multi-lane freeways. particularly in mountainous regions or in river gorges and canyons.1 are commonly used depending on the conditions at the site. In some instances. As a consequence. Piers. river valleys provide the only feasible route for highways.1 Types of Encroachment In the vicinity of rivers.2 . and pattern. channel bed conditions. In some instances. this encroachment becomes severe. 1. eccentricity. Floodplains often appear to provide an attractive low cost alternative for highway location. reducing the required bridge length. the economics of crossings require substantial encroachment on the river and its floodplain. a stretch of the river must be straightened to eliminate meanders to accommodate the highway. submergence of the superstructure. ice. and those involved in transportation. The approaches may be skewed or normal (perpendicular) to the direction of flow or one approach may be longer than the other. highways. and flood control mistakenly consider a river to be static. or in the form of piers and abutments or culverts in the main channel of the river.2 Geometry of Bridge Crossings The bridge crossing is the most common type of river encroachment. In some regions. The design procedures in this document have been derived from laboratory and field observations of bridge crossings. including interchanges. More commonly. These design procedures take advantage of the large volume of work that has been done by many people in describing the hydraulics and scour characteristics of bridge crossings. navigation.1. Longitudinal encroachments may exist that are not connected with river crossings. or the abutments may extend up to the banks or even protrude over the banks. The geometric properties of bridge crossings illustrated in Figure 1. The design procedures include allowances made for the effects of skewness. an alluvial river generally is continually changing its position and shape as a 1.2 DYNAMICS OF NATURAL RIVERS AND THEIR TRIBUTARIES Frequently. This is true even in areas where a floodplain does not exist. 1. The encroachment can be in the form of earth fill embankments on the floodplain or into the main channel itself. often encroach on a floodplain over long distances. hydraulic engineers. producing an eccentric crossing. river crossings can be accomplished with absolutely no encroachment on the river. Abutments used for the overbank-flow case may be set back from the low-flow channel banks to provide room to pass the flood flow or simply to allow passage of livestock and machinery. spurs.1. the cost of a single span over the entire floodplain being prohibitive. wind waves. debris. that is. However. The bridge and its approaches are located far above and beyond any possible flood stage. particularly as older highways are upgraded and widened. spurs and guide banks add to the list of geometric classifications.

and the past activities of man (archaeologists). slopes. Rivers. glaciers.2. rarely consider the landscape as unchanging. sand dunes. scientists concerned with the history of landforms (geomorphologists). all rivers are governed by the same basic forces. The design engineer must understand. The response of a river to human-induced changes often occurs in spite of attempts by engineers to keep the anticipated response under control. Over a relatively short period of time.1 Historical Evidence of the Natural Instability of Fluvial Systems In order to emphasize the inherent dynamic qualities of river channels. including the probable geometric alterations that will be activated by the changes a project and future projects will impose on the channel Hydraulic characteristics such as depths. 1. and seacoasts are highly susceptible to change with time. When an engineer modifies a river channel locally. including possible changes in flows.consequence of hydraulic forces acting on its bed and banks. evidence is cited below to demonstrate that most alluvial rivers are not static in their natural state. Geometric properties of bridge crossings. In spite of their complexity. and the hydrologic effects of changes in land use Geometric characteristics of the stream. perhaps in some cases up to 100 years.1. Indeed. this local change frequently causes modification of channel characteristics both up and down the stream. components of the 1. These changes may be slow or rapid and may result from natural environmental changes or from changes by human activities. and velocity of streams and what changes may be expected in these characteristics in space and time. Figure 1. The points that must be stressed are that a river through time is dynamic and that human-induced change frequently sets in motion a response that can be propagated upstream or downstream for long distances. including soil conditions Hydrologic factors. and work with these natural forces: • • • • Geological factors. runoff.3 . vegetation (botanists).

and this reduction of baselevel caused major incisions of river valleys near the coasts. shape.2 a section of the Mississippi River as it was in 1884 is compared with the same section as observed in 1968. and pattern. and because of its dimensions it has sometimes been considered unique. In the lower 9. During the last continental glaciation. In Figure 1. Equally significant during this time were fluctuations of sea level. pattern) as a result of changes of hydrology. This. Nevertheless. in fact. or as a result of natural stochastic climatic fluctuations (droughts. Changes may be very slow or dramatically rapid. The Mississippi may be thought of as a prototype of many rivers or as a much larger than prototype model of many sandbed rivers. of course. Missouri. it is the rule rather than the exception that banks will erode. 1. dimensions. Rivers are.4 . the major climatic changes of recent geological time (the last few million years of earth history) have triggered dramatic changes in runoff and sediment loads with corresponding channel alteration. Hydrology can change as a result of climatic changes over long periods of time. or by human modification of the hydrologic regime. shape. The Mississippi is our largest and most impressive river.landscape may be relatively stable.6 km (6 mi) of river. floods). islands. For example. sea level was on the order of 120 m (400 ft) lower than at present.2. Some of this change has been natural and some has been the consequence of river development work. Rivers change position and morphology (dimensions. Evidence from several sources demonstrates that river channels are continually undergoing changes of position. sediments will be deposited and floodplains. Comparison of the 1884 and 1968 Mississippi River Channel near Commerce. Fisk's (1944) report on the Mississippi River and his maps showing river position through time are sufficient to convince everyone of the innate instability of the Mississippi River. Hydraulic and geomorphic laws apply at all scales of comparable landform evolution. the surface area has been reduced approximately 50 percent during this 84-year period. stability cannot be automatically assumed. In alluvial river systems. the most actively changing of all geomorphic forms. and side channels will undergo modification with time. is not so. Figure 1.

A compilation of data by Wolman and Leopold shows that rates of lateral migration for the Kosi River of India range up to approximately 760 m/yr (2. Wallace (1967) has shown that many small streams are clearly offset laterally along the San Andreas Fault in California. the progressive shifting of a river toward one side of its valley has resulted from lateral tilting. changes of channel geometry as a result of climatic and hydrologic changes.5 . the geologist is not surprised to see drainage patterns that have been disrupted by uplift or some complex warping of the earth's surface. Of course. rivers from the geomorphic point of view are unquestionably dynamic. permanence of vegetation on banks and the floodplain and watershed land use. For example. may also be causes of modern river instability. convexities in the longitudinal profile of both rivers and river terraces (these profiles are concave under normal development) have been detected and attributed to upwarping. The movement of the earth's crust is one geologic agent causing modern river instability. Further. The earth's surface in many parts of the world is undergoing continuous measurable change by upwarping. complete reversals of drainage lines have been documented. The history of semi-arid and arid valleys of the western United States is one of alternating periods of channel incision and arroyo formation followed by deposition and valley stability which have been attributed to climatic fluctuations. Rates of lateral migration for two major rivers in the United States are as follows: Colorado River near Needles. which are interesting from an academic point of view. the study of these ongoing changes (called neotectonics) has become a field of major interest for many geologists and geophysicists. subsidence or lateral displacement. Much. the shift of a channel to form chutes and islands. Such gradual surface changes can affect stream channels dramatically. Progressive lateral movement of this fault on the order of 25 mm (an inch) per year has been measured.100-year design life of a bridge is considered. Climatic change. Long-term climatic fluctuations have caused major changes of river morphology. that is. and obliteration or displacement of existing channels by continental glaciation. neotectonics should not be ignored as a possible cause of local river instability. Major shifts in position of the Brahmaputra River toward the west are attributed by Coleman (1969) to tectonic movements. These types are deep incision and deposition as sea level fluctuated. In fact.6 mm (0. depends on flood events. therefore.000 years. this rate is actually 7. Seemingly insignificant in human terms. sea level change. Hence. the slope of the energy gradient on the Lower Mississippi River is about 47 to 95 mm/km (3 to 6 in/mi). and glaciation. bank stability. a river may maintain a stable position for long periods and then experience rapid movement. For many river systems. and the cutoff of a bend to form oxbow lakes. neither neotectonics nor a progressive climate change will have a detectable influence on river character and behavior. Hence. particularly when the 50. California. What then causes a river to appear relatively unstable from the point of view of the highway engineer or the environmentalist? It is the slow but implacable shift of a river channel through erosion and deposition at bends. In addition. 3 1. a change of slope of 76 mm (3 in) would be significant. but does this apply to modern rivers? It is probable that during a period of several years. major river changes of different types occurred. As a result. Lateral migration rates are highly variable. It is clear that rivers can display a remarkable propensity for change of position and morphology in time periods of a century. The rates of movement of faults are highly variable. For example. but an average rate of mountain building has been estimated by Schumm (1963) to be on the order of 7.500 ft/yr).In recent geologic time. Floodplains have been destroyed and reconstructed many times over.6 m (25 ft) per 1.3 in) per year or 76 mm (3 in) per decade.

1. decrease in annual flow by irrigation and restriction of channel width by bridges. working on the Rio Ycayali in the Amazon headwaters of Peru. it was at times 305 m/yr (1. The island formed in a location protected from the erosive effects of floods but susceptible to deposition of sediment during floods. Mississippi. Equally dramatic changes of channel dimensions have occurred along the North and South Platte Rivers in Nebraska and Colorado as a result of control of flood peaks by reservoir construction. On the other hand. Kentucky. which was 15 m (50 ft) wide during the latter part of the 19th and first part of the 20th centuries (Schumm and Lichty 1957).000 ft) wide but now are less than 300 m (1. On the Lower Mississippi River.200 ft) and the channel occupied essentially the entire valley floor. For example. a meander loop pattern of the lower Ohio River has altered very little during the past thousand years (Alexander and Nunnally 1972). a barge became stranded in a chute of the Mississippi River near Columbus. A study by Schmuddle (1963) shows that about one-third of the floodplain of the Missouri River over the 274 km (170 mi) reach between Glasgow and St. which testifies to the rapidity of fluvial processes. In their natural state the rivers were 600 to 1. Lathrop (1968). 1944). bend migration was on the order of 0. below Cairo. Although the dynamic behavior of perennial streams is impressive.000 years. Mississippi River near Rosedale. Archaeologists have also provided clear evidence of channel changes that are completely natural and to be expected. During the decade of the 1940s a new floodplain was constructed. was reworked by the river between 1879 and 1930. These loops have an amplitude of 3. the number of archaeologic sites of the floodplains decreases significantly with age because the earliest sites are destroyed as floodplains are modified by river migration.000 ft). 48 to 192 m/yr (158 to 630 ft/yr). whereas in the central and upper parts of the river. the side channel separating the island from the mainland had filled to the extent that the island became part of Missouri. and the river width was reduced to about 152 m (500 ft) in 1960. For these reasons the channel filling was rapid and progressive.7 km (2 to 6 mi) and an average rate of meander growth of approximately 12 m/yr (40 ft/yr). the modification of rivers in arid and semi-arid regions and especially of ephemeral (flowing occasionally) stream channels is startling.2 to 9. In 1919. These topics are discussed in detail in Chapter 5. and a few acres were cleared for agricultural purposes. The barge induced deposition in the chute and an island formed. Missouri. Changes of this magnitude due to changes in flow are perhaps exceptional.6 m/yr (2 ft/yr). As exceptional example of this is the Cimarron River of Southwestern Kansas. Historical and field studies by Smith (1940) show that floodplain destruction occurred during major floods on rivers of the Great Plains. but emphasize the mobility of rivers and their ability to adapt to changing conditions.6 . It cannot be concluded that islands will always form and side channels fill at such rapid rates.to 46 m/yr (10 to 150 ft/yr).500 m (2000 to 5. Another somewhat different type of channel modification. Following a series of major floods during the 1930s it widened to 366 m (1. is described by Shull (1922. A study of floodplain vegetation and the distribution of trees in different age groups led Everitt (1968) to the conclusion that about half of the Little Missouri River floodplain in western North Dakota was reworked in 69 years.000 ft/yr) (Kolb 1963). estimates that on the average a meander loop on this river begins to form and cuts off in 5. Charles. the island was sufficiently large to be homesteaded. During a major flood in 1913. By 1933. but island formation and side-channel filling appear to be the normal course of events in any river transporting moderate or high sediment loads regardless of the river size.

With dunes. This topic is discussed in detail in Chapter 3. discharge.3. The mechanics of flow in rivers is a complex subject that requires special study. In summary. then a long period of channel instability can be envisioned with considerable bank erosion and lateral shifting of the channel before stability is restored. if changes in sinuosity and meander wavelength as well as in width and depth are required to compensate for a hydrologic change. Such a change could be caused by a natural or artificial cutoff.2 Introduction to River Hydraulics and River Response In the previous section. major adjustments of channel morphology can be anticipated. it is conceivable that a decrease in discharge together with an increase in sediment load could actuate a decrease in depth and an increase in width. botanical. a few simple hydraulic and geomorphic concepts are presented here. is illustrated schematically in Figure 1. At low flow or with warm water the bed of a sand bed stream can be dunes. 1. archaeological. The significantly different channel dimensions. 1. Whereas. shapes. bars and forms of bed roughness with changing water and sediment discharge. and patterns associated with different quantities of discharge and amounts of sediment load indicate that as these independent variables change. is subdivided by sandbars and carries relatively large quantities of sediment. but any changes that are imposed on a river may change its form. which may be imposed independent of the other river characteristics. For example. it is possible to change the river from a meandering one that is relatively tranquil and easy to control to a braided one that varies rapidly with time. The reaction of a channel to changes in discharge and sediment load may result in channel dimension changes contrary to those indicated by many regime equations.2. but at large flows or cold temperature the bed may become plane or have antidune flow. the bed configuration of a river can change with temperature. The direction and extent of the change depends on the forces acting on the system. The dependence of river sinuosity on the slope.In addition to changes in planform location and size with time. The major complicating factors in river mechanics are: (a) the large number of interrelated variables that can simultaneously respond to natural or imposed changes in a river system and (b) the continual evolution of river channel patterns.7 . geological. and concentration of silts and clays. with plane bed or antidune flow the resistance to flow is small and the bed material transport is large. which is unfortunately not included in basic courses of fluid mechanics. and geomorphic evidence supports the conclusion that most rivers are subject to constant change as a normal part of their morphologic evolution. resistance to flow is large and bed material transport is low. channel geometry. Therefore. By changing the slope. Conversely. River forms are broadly classified as straight. Further. braided or some combination of these classifications. it was established that rivers are dynamic and respond to changing environmental conditions. has high velocities. In order to understand the responses of a river to the actions of humans and nature. stable or static channels are the exception in nature. meandering. it is possible that a slight decrease in slope could change an unstable braided river into a meandering one.

where major tributaries exert a significant influence on the main channel by introducing large quantities of sediment. It is imperative that anyone working with rivers.8 . have an understanding of the many factors involved. upstream control on the main channel may allow the tributary to intermittently dominate the system with deleterious results. Prior to this interest in environmental impacts. hydrology. in fact. and of the potential for change existing in the river system. For example. Changes in sediment and water discharge at a particular point or reach in a stream may have an effect ranging from some distance upstream to a point downstream where the hydraulic and geometric conditions will have absorbed the change. flood control structures can cause downstream flood damage to be greater at reduced flows if the average hydrologic regime is changed so that the channel dimensions are actually reduced. slope with constant discharge. Artificial controls that could benefit the reach may. Sinuosity vs. it is necessary to consider a channel reach as part of a complete drainage system. either with localized areas or entire systems. soils.Figure 1. cause problems in the system as a whole. Thus. very few people interested in rivers ever considered the long-term changes that were possible. sediments from the tributary that previously were eroded will no longer be carried away and serious aggradation with accompanying flood problems may arise.3. 1. If discharges in the main channel are reduced. and the time involved to reach a new equilibrium can be gained by: • • • • Studying the river in a natural condition Having knowledge of the sediment and water discharge Predicting the effects and magnitude of future human activities Applying knowledge of geology. the magnitude of change. and hydraulics of alluvial rivers The current interest in ecology and the environment have made people aware of the many problems that humans can cause. An insight into the direction of change. Also.

As a consequence of construction.3. With proper awareness of the large-scale effects that can exist. To study a transient phenomenon in natural alluvial channels. The transported sediment is carried from the construction site by surface flow into the minor rills. the aesthetics of the river environment and short-term and long-term effects of erosion and sedimentation on the surrounding landscape and the river. the equations of motion and continuity for sediment laden water and the continuity equation for sediment can be used as discussed in Chapters 2. the river may be capable of carrying the increased sediment load an additional distance. the development of crossings and the contraction of river sections may have a significant effect on the water level in the vicinity and upstream of the bridge.3 EFFECTS OF HIGHWAY CONSTRUCTION ON RIVER SYSTEMS Highway construction can have significant general and local effects on the geomorphology and hydraulics of river systems. but only until a reduction in gradient and a reduction in transport capability is encountered. The water 1. including the river's aesthetic value. Hence. Engineers have long used small-scale hydraulic models to assist them in anticipating the effect of altering conditions in a reach of a river. channel stabilization. Such changes in water level upstream of the bridge are called backwater effects. the results of hydraulic model testing can be extremely useful for this purpose. which combine within a short distance to form larger channels leading to the river. and countermeasures. In many instances. every effort should be made to accomplish the channelization in a manner which does not degrade the river environment. it is necessary to consider induced short-term and long-term responses of the river and its tributaries. contraction due to the construction of encroachments usually cause contraction and local scour.Two methods of predicting response are physical and mathematical models. In addition. to offset increased velocities and to reduce bank instabilities and related problems. An alternative method of predicting short-term and long-term changes in rivers involves the use of mathematical computer models. the river is stabilized or channelized to some degree. The highway engineer must be in a position to accurately assess the effects of the construction of crossings upon the water surface profile. the impact on environmental factors. 4. many areas become highly susceptible to erosion. The biological response of the river system should also be considered and evaluated. 3. In the event that the contraction is extended further downstream.9 .1 Immediate Response of Rivers to Encroachment Let us consider a few of the numerous and immediate responses of rivers to the construction of bridges. Local changes made in the geometry or the hydraulic properties of the river might be of such a magnitude as to have an immediate impact upon the entire river system. 1. When it is necessary to do this. and 5. 1. The increased velocities caused by encroachments may also affect the general lateral stability of the river downstream. At bridges. and the sediments removed from this location are usually dropped in the immediate reach downstream.

The large sediment particles transported to the main channel may remain in the vicinity of the construction site for a long period of time or may be slowly moved away. increasing the flow velocities and possibly causing instability in the river at that site. The surface runoff and the accompanying erosion can significantly increase the sediment yield to the river channel unless careful control is exercised. Certain species of fish can only tolerate large quantities of suspended sediment for relatively short periods of time. biology.3. Along the same lines. However. construction work within the river may cause a loss of food essential to fish life and often it is difficult to get the food chain reestablished in the system. the engineer should utilize adequate technical assistance from experts in fisheries. spurs. This type of biological response to development normally falls outside of the competence of the engineer. Runoff from highway construction may increase sediment and turbidity of the water.2 Delayed Response of Rivers to Encroachment In addition to the examples of possible immediate response discussed above. they are responses that illustrate the importance of considering the environment in the design of highway encroachments. As part of this introductory 1. there may be an immediate drop in the quality of the fishing due to the increase of sediment load or other changed hydraulic characteristics within the channel. The introduction of larger quantities of sediment into the channel and changes made in the geometry of the channel may result in the loss of these pools and riffles. This is particularly true of the eggs and fry. construction of barbs. In many streams. Both form an important part of the environment for fish. the fine sediments are easily transported and generally disperse into the whole cross section of the river. engineering works may be responsible for the discharge of these sediments into the system. Yet. Another possible immediate response of the river system to construction is the loss of the recreational use of the river. In contrast. there are important delayed responses of rivers to highway development. Construction and operation of highways in water supply watersheds present very real problems and may require special precautionary designs to protect the water supplies from highway runoff or accidental spills.flowing from the construction site is usually a consequence of rain. provide a better habitat for fish and in general improve the environment.10 . thereby steepening the channel. Only with such knowledge can one develop the necessary arguments to "sell" the case that erosion control measures must be exercised to avoid significant deterioration of the stream environment not only in the immediate vicinity of the bridge but in many instances for great distances downstream. The preceding discussion is related to only a few immediate responses to construction along a river. and other related areas to assess the consequences of sediment in a river. On the other hand. the sudden injection of the larger sediments into the channel may cause local aggradation. In this regard. Some natural rivers consist of a series of pools and riffles. The fine sediments are transported downstream to the nearest reservoir or to the sea. The suspended fine sediments can have very significant effects on the biomass of the stream. As will be discussed later. 1. and other river control works to stabilize the river channel and protect a bridge crossing can decrease erosion.

It is essential to consider the probable long-term plans of all agencies and groups as they pertain to a river when designing crossings or when dealing with the river in any way. thereby causing aggradation in the tributaries. 1. the sediment drops out rapidly. for example. It is necessary also to consider the effects on highway encroachments of river development works. Sometimes it is necessary to employ training works in connection with highway encroachments to align the flow with bridge or culvert openings. culverts. The general consequence of cutoffs is to shorten the flow path and steepen the gradient of the channel. degradation can be induced of such magnitude as to cause failure of structures such as bridges. Any such changes made in the system that cause an increase in the gradient may cause an increase in local velocities. The increase in velocity increases local and contraction scour with subsequent deposition downstream where the channel takes on its normal characteristics. Having a lower water level in the main channel for a given discharge means that the tributary streams entering in that vicinity are subjected to a steeper gradient and higher velocities which cause degradation in the tributary streams. shorten the flow line. Deposition can have significant detrimental effect on the downstream reach of river. and the utilization of training works have been mentioned. channel stabilization. they generally straighten the channel. any increase in transported materials from the tributaries to the main channel causes a reduction in the quality of the environment within the river. In this region of slower velocities. If significant lengths of the river are trained and straightened. consideration is given to some of the more obvious effects that can be induced on a river system over a long time period by highway construction.chapter. water diversions to and from the river system. this action can induce significant instability such as bank erosion and degradation in the reach. approaches. The local steepening can significantly increase the velocities and sediment transport. small towns had developed and small tributary streams entered the main channel within the bendway. increasing the flood stage in the river itself and increasing the base level for the tributary stream. and the mining of sand and gravel. Cutoffs may develop naturally in the river system or they may be the result of human activity. Let us consider a few typical responses of a crossing to different types of water resources development. It was decided to develop a cutoff 1. Along this bend. Increased sediment loads usually result in a deterioration of the environment. When such training works are used. bank instabilities are induced and the sediment loads are greatly increased. In general. construction of reservoirs. navigation works. cutoffs. These works may include. longitudinal encroachments. In extreme cases. there can be a noticeable decrease in the elevation of the water surface profile for a given discharge in the main channel.11 . Also. culverts or other encroachments on the tributary systems. and increase the local velocity within the channel. Consider a classic example of a cutoff that was constructed on a large bend in one of the tributaries to the Mississippi. Tributaries emptying into the main channel in such reaches are significantly affected.4 EFFECTS OF RIVER DEVELOPMENT ON HIGHWAY ENCROACHMENTS Some of the possible immediate and delayed responses of rivers and river systems to the construction of bridges. levees. flood control works. More specifically. as degradation occurs in the tributaries. The material scoured in the reach affected by the cutoff is probably carried only to an adjacent downstream reach where the gradient is flatter.

The normal stage in that channel was increased by a factor of 4 to 5. it must be recognized that downstream of storage structures the channel may either aggrade or degrade and the tributaries will be affected in either case. In such a haphazard program of river development. The banks were nicely vegetated. its sediment load is deposited forming a delta.12 . reduce the flood stage and generally improve poor hydraulic conditions in that location. changing the channel to a straight raging torrent capable of carrying large 1. This clear water has the capacity to transport more sediment than is immediately available. the additional flow gutted the river valley. the small tributaries continued to discharge their water and sediment. Consequently the channel begins to supply this deficit with resulting degradation of the bed or banks. Ultimately the river may be subjected to a flow of magnitude sufficient to overflow existing banks. Within a short period of time. Years ago.across the meander to shorten the flow line of the river. When the stream flowing into a reservoir encounters the ponded water. The clear-water diversion into South Boulder Creek in Colorado is another example of river development that affects bridge crossings and encroachments as well as the environment in general. The flattening of the upstream channel induces aggradation causing the bed of the river to rise. however that the additional sediments supplied by the tributary streams would ultimately offset the degradation in the main channel. causing the water to seek an entirely new channel. the highway engineer would be hard pressed to maintain and plan for the highway system along and over this reach of river. essentially clear water is released downstream of the dam site. This degradation may significantly affect the safety of bridges in the immediate vicinity. The extra water caused significant bank erosion and channel degradation. For example. In fact. has caused the Rio Grande to aggrade many miles upstream of the reservoir site. Within the bendway. This deposition in the reservoir flattens the gradient of the channel upstream. this channel section could not convey the sediment from these small stream systems and aggradation was initiated. As a consequence of the adverse action in the vicinity of the cutoff and within the bendway itself. Elephant Butte Reservoir. other hydraulic structures and all types of training and stabilization works. sufficient aggradation had occurred so as to jeopardize water intakes and sewage outfalls. the degraded or widened main channel causes steeper gradients on tributary streams in the vicinity of the main channel. Because of the flat gradient in the bend. Several interesting results developed. water was diverted from the Western Slope of the Rockies through a tunnel to the North Fork of South Boulder Creek. threatening highway installations and other facilities. With sediment trapped in the reservoir. With the abandonment of the existing channel there would be a variety of bridges and hydraulic structures that would also be abandoned at great expense to the public. Originally the North Fork of South Boulder Creek was a small but beautiful scenic mountain stream. there was a beautiful sequence of riffles and pools. the bankline eroded and degradation was initiated. These reservoirs serve as traps for the sediment normally flowing through the river system. This change in bed level can have very significant effects upon bridges. which had all the attributes of a good fishing habitat. Significant responses can be induced upstream of reservoirs as well as downstream. Thus. it was finally decided that it would be more beneficial to restore the river to its natural form through the bend. Another common case occurs with the development of reservoirs for storage and flood control. built on the Rio Grande in New Mexico. It is entirely possible. The result is degradation in the tributary streams. In the vicinity of the cutoff. Again. This action was taken and the serious problems were alleviated.

Hoover Mackin (1937) when he wrote: "The engineer who alters natural equilibrium relations by diversion or damming or channel improvement measures will often find that he has the bull by the tail and is unable to let go.quantities of sediment. . as he continues to correct or suppress undesirable phases of the chain reaction of the stream to the initial 'stress' he will necessarily place increasing emphasis on study of the genetic aspects of the equilibrium in order that he may work with rivers. an analytical approach to the problem can be very difficult and time consuming. The valley may never again have the natural form and beauty it once possessed. Stabilization was achieved by flattening the gradient by constructing numerous drop structures and by reforming the banks with riprap. beauty and utility. In the preceding paragraphs. . Degradation in the system had reached as much as 5 to 6 m (15 to 20 ft) before measures were taken to stabilize the creek. rather than merely on them. . and the natural sequence of riffles and pools has been destroyed. Attempts at controlling large rivers have often led to the situation described by J. . This important aspect of the response of rivers to development is treated in detail in later chapters. if a greater understanding of the principles governing the processes of river formation is to be gained. We should bear in mind that diversions to or from the natural river system can greatly alter its geometry. Nevertheless. to prevent or reduce the detrimental effects of any modification of the natural processes and state of equilibrium on a river.13 . The system has stabilized but it is a different system. The channel is straight.5 TECHNICAL ASPECTS Effects of river development." Through such experiences. one realizes that. but no guidance has been given on how to determine the magnitude of these changes. flood control measures and channel structures built during the last century have proven the need for considering delayed and far-reaching effects of any alteration humans make in a natural alluvial river system. The river may undergo a complete change. encroachments and bridge crossings along the affected reach. one must gain an understanding of the physical laws governing them and become knowledgeable of the far-reaching effects of any attempt to control or modify a river's course. . Because of the complexity of the processes occurring with natural channel flows and the accompanying erosion and deposition of material. giving rise to a multitude of problems in connection with the design and maintenance of hydraulic structures. possible immediate and long-term responses of river systems to various types of river development have been described. much of the vegetation has been washed away. Most of our river process relations have been derived empirically. 1. the empirically derived relations must be put in the proper context by employing an analytical approach. 1.

quite difficult to relate mathematically.1. structures. One approach is to list measurable or computable variables which effectively describe the processes occurring and then to reduce the list by making simplifying assumptions and examining relative magnitudes of variables. statistician.5. 1. the highway engineer must further improve his or her knowledge. and (7) human activity. geologist. In addition. In fact. as the public demands more comprehensive treatment of river development problems. the environment and related subjects. erosion and sedimentation.14 . (5) vegetation. maintenance. The fluvial processes involved are very complicated and the variables of importance are difficult to isolate. hydrologist.3 Data Requirements Large amounts of data pertaining to understanding the behavior of rivers have been acquired over a long period of time. viscosity. quantify their interrelations where feasible. it is extremely difficult to study the role of any individual variable.1 Variables Affecting River Behavior Variables affecting alluvial river channels are numerous and interrelated. professional organizations representing these disciplines should be encouraged to work cooperatively to achieve long-range research needs and goals relative to river development. and show how they can be applied to achieve the successful design of river crossings and encroachments. Significant variables are. the basic equations of fluid motion may be simplified (on the basis of valid assumptions) to describe the physical processes within an acceptable balance between accuracy and limitations of obtaining data. including types of sediments. define them. soil chemist. meteorologist. The problem has been more amenable to an empirical solution than an analytical one. 1. hydraulics.5. soil physicist. computer programmer. river mechanics. water management staff and economist. soil mechanics. Nevertheless.5. It is the role of the succeeding chapters to present these variables. In all aspects of bridge inspection. geomorphologist. some data collection efforts have been sporadic and unfocused. (6) geology. In an analysis of flow in alluvial rivers. show how they interrelate. When this is done. (2) sediment discharge.2 Basic Knowledge Required In order for engineers to cope successfully with river engineering problems. mathematician. design and construction an interdisciplinary approach is needed. unlike rigid boundary hydraulic problems. systems engineer. the flow field is complicated by the constantly changing discharge. by soliciting the cooperative efforts of the hydraulic engineer. temperature. Many laboratory and field studies have been carried out in an attempt to relate these and other variables. (4) bank and bed resistance to flow. Their nature is such that. Agencies should take a careful look at present data requirements needed to solve 1. and the application of it. it is necessary to have an adequate background in engineering with an emphasis on hydrology. biologist. Major factors affecting alluvial stream channel forms are: (1) stream discharge. (3) longitudinal slope. economics. therefore.

gov/bridge/hydsoft. if not all. There is good understanding of stable channel theory in non-cohesive materials of all sizes. Environmental data is also needed so that proper assessment can be made of the impact of river development upon the environment and vice versa.usace.fhwa.practical problems along with existing data. The steady-state sediment transport of nearly uniform sizes of sediment in alluvial channels is well understood. There have been extensive studies of the fall velocity of non-cohesive sediments in static fluids to provide knowledge about the interaction between the particle and fluid so essential to the development of sediment transport theories. designs can be made for appropriate types of stabilization treatments so that canals and rivers behave in a stable manner. Advances have been made in developing computational software to establish hydraulic variables for scour computations. training programs and methods of developing staff that can apply this knowledge to the solution of practical problems.mil/software/index.6 FUTURE TECHNICAL TRENDS When considering the future.army. In order to correct such deficits there is a need to take a careful look at existing data pertaining to rivers. and encourage further research to help correct these deficits of knowledge.dot. The problem of data requirements at river crossings is of sufficient importance that it is treated in greater detail in Chapter 8. it is essential to recognize the present state of knowledge pertaining to river hydraulics and then identify inadequacies in existing theories and our understanding of the physical processes involved. A careful analysis of data requirements would make it possible to more efficiently utilize funds to collect data in the future. of the commonly used scour prediction equations have been incorporated into these models. steady and unsteady models.hec. Most. However. and the characteristics of watersheds and how they deliver water and sediment to the stream systems. including 1. the characteristics of the sediments being transported by streams. research needs. sediment discharge hydrograph.6. 1. the characteristics of the channels in which the water and sediment are transported. applications methodologies are required to facilitate the use of more appropriate hydraulic variables that can be obtained from more sophisticated computer models.and 2-dimensional.15 . momentum and energy concepts are well known and can be effectively applied to a wide variety of river problems. if necessary. Considerable work has been done on the hydraulics of rigid boundary open channels and excellent results can be expected if the principles are applied properly.htm + www. World wide web sites providing hydraulic models applicable to scour computations include: + www.1 Adequacy of Current Knowledge The basic principles of fluid mechanics involving application of continuity. future data requirements.html 1. The theory is adequate to enable us to design stable systems in the existing bed and bank material or. The basic type of information that is required includes: water discharge hydrograph. 1.

devote only a small amount of time to teaching hydrology. On the other hand. However. These interfaces are used for model development. and related problems.The use of computers and the development of computer programs have greatly helped the hydraulic engineer to solve problems on highway crossings and encroachments. The level of effort required to apply these models to bridge and hydraulic structure design has been greatly reduced with the availability of graphical user's interfaces. a more basic theoretical understanding of flow in the river systems needs to be developed. FHWA’s BRI-STARS program can be used to determine flow of water and sediment through a bridge crossing. which can be applied to the behavior of rivers. Some research needs are particularly urgent and promise a rather quick return. troubleshooting. the study of bed forms generated by the interaction between the water and sediment in the river systems deserves further study. the number of individuals who are cognizant of existing theory and can apply it successfully to the solution of river problems is limited. Simple terms have been used to describe the characteristics of alluvial material of both cohesive and non-cohesive types. two-dimensional velocity distributions. river mechanics. Undergraduate engineering educators in the universities in the United States. Thus. a comprehensive look at the characteristics of materials is warranted. are extensive. Both the Corps' and HEC-RAS (Version 3x) and UNET models are particularly useful in unsteady flow problems. Stabilization of rivers and bank stability of river systems needs further consideration. There is a specific reason for this deficit of trained personnel. The Corps of Engineers HEC 1 and related HEC HMS programs are of particular importance in routing water from an upstream gaging station downstream to a bridge. in many instances only empirical relationships have been developed and these are pertinent to specific problems only. There is great need for adequately trained and experienced practitioners to cope with river problems.16 . 1. and only a limited number of universities and institutions offer the required training in these subject areas. Also. and in the world for that matter. but theories pertaining to their development are inadequate. fluvial geomorphology. and review. secondary currents. The one-dimensional WSPRO (FHWA) or HEC-RAS (Corps of Engineers) computer programs are extremely useful to determine the water surface elevation and the velocity of the flow in a river at a highway crossing.2 Research Needs As knowledge of river hydraulics is reviewed. available concepts and theories. channel stabilization. fall velocity of particles in turbulent flow and the application of remote sensing and geographic information system techniques to 1. output presentation. Consequently. it becomes quite obvious that many things are not adequately known. Particularly. With respect to many aspects of river mechanics. It is not possible to obtain adequate training in these important topics except at the graduate level. FHWA’s FESWMS and the Corps of Engineers RMA-2V 2-dimensional models provide water surface elevation and local velocities in two dimensions for unsteady flows for complex problems such as tidal flows. taking into account any increase or decrease in flow that might occur in between the station and the bridge.6. the number of individuals involved in the actual solution of applied river mechanics problems is very small. Other important research problems include the fluid mechanics of the motion of particles. it can be concluded that knowledge is available to cope with the majority of river problems. The types of bed forms have been identified.

17 . A comprehensive study on information theory is needed to cope with such difficulties. flow pattern and pressure distribution within the flow depend on gravity. no deformation of the bed and banks is considered. 1.1. Operational research on decision making. river hydraulics. particularly at the undergraduate level. is another pressing research area. 1. engineering training is often inadequate in relation to understanding the dynamics of rivers. Insufficient data is frequently a problem of river mechanics analysis.6. mathematical modeling of water and sediment yield from small watersheds. embankments and at river control structures.Open Channel Flow This chapter describes the fundamentals of rigid boundary open channel flow. surface configuration. In rigid boundary open channel flow. environmental considerations and related topics. Laboratory demonstrations are an inexpensive method of quickly and effectively teaching the fundamentals of river mechanics and illustrating the behavior of structures. In addition.3 Training As pointed out in Section 1. Studies are needed to better define the planform of rivers and their response to changing flow conditions. Mobile 1. These demonstrations should be followed by field trips to illustrate similarities and differences between phenomena in the laboratory and in the field. mathematical modeling of river response followed by field verification. Better ways to train engineers and to disseminate existing knowledge in this important area need to be considered.1 Chapter 2 . which can generate deformation of the boundary through scour and fill. erosion and sedimentation. the water surface is not confined. A primary research need is the collection of field data on the flow variables and depth of scour at bridges. The physical modeling of rivers followed by prototype verification. Finally the results of these efforts must be presented in such a form that it can be easily taught and easily put to practical use. considering cost and risk criteria to determine the hydrologic and hydraulic design of highway structures and project alternatives. laboratory and field studies are needed to improve the equations for estimating total scour at piers and abutments. Mobile boundary hydraulics refers to flow. At the very minimum. 1. In open channel flow.7.hydrology and river mechanics. The curriculum of university education made available to engineers should be improved. such a curriculum should strive to introduce concepts of fluvial geomorphology. Formal training should be supported with field trips and laboratory demonstrations. and studies of unsteady sediment transport are areas in which significant advances can be made.7 OVERVIEW OF MANUAL CONTENTS In the following sections a brief overview of each chapter will be given.6.

The amount of sediment transported or deposited in a stream under a given set of conditions is the result of the interaction of two groups of variables. Silt generally is not present in appreciable quantities in streams having non-cohesive boundaries. and 5.boundary hydraulics will be discussed in Chapters 3. 1. and presents equations and computer models that other practitioners have developed. gravel. presents equations. Chapter 4 presents the terms that describe sediment transport. and can still be. presents methods for prediction of bed forms.18 . These fundamentals of alluvial channel flow are used in later chapters to develop design considerations for highway crossings and encroachments in river environments. laminar or turbulent. the rivers are formed in cohesive or non-cohesive materials that have been. where velocity and acceleration are large only in one direction and are so small as to be negligible in all other directions.3 Chapter 4 . and describes the methods used for the physical measurement and calculation of sediment discharge. 4. and momentum are derived and used. derives and explains the basic equations. 1. The non-cohesive material generally consists of silt. sand. transported by the stream. Streambed erosion and movement is the result of sediment transport.004 mm) forming a binder with silts and sand Chapter 3 presents the fundamentals of alluvial channel flow. tables and figures for determining the beginning of motion of non-cohesive sediments. and tranquil or rapid (subcritical or supercritical) open channel flow are described. Chapter 2 is restricted to a 1-dimensional analysis of rigid boundary open channel flow. Example problems are solved at the end of the chapter in both SI and English units of measurement. That is. The three basic equations of flow: continuity. Cohesive material consists of clays (sizes less than 0. Example problems are solved at the end of the chapter in both SI and English units of measurement.7. energy. The characteristics of uniform or nonuniform. describes flow in sandbed channels and associated bed forms. Example problems to determine the quantity and quality of sediment transport are solved at the end of the chapter in SI and English units of measurement. methods of measuring these properties. Scour at a bridge or culvert is a sediment transport process. In the second group are the variables that determine the capacity of the stream to transport that sediment. discusses the methods and fluid forces that move sediment. In the first group are those variables that determine the quantity and quality of the sediment brought down to that section of the stream (bed and bank material).2 Chapter 3 . It covers properties of alluvial material. and Manning's n for sandbed and other natural streams.Fundamentals of Alluvial Channel Flow Most streams that a highway will cross or encroach upon are alluvial. or cobbles. equations derived and problems for highways in the river environment are solved.7.Sediment Transport The quantity and quality of the sediments that a stream can transport is an important consideration for highways in the river environment. or any combination of these sizes. steady or unsteady. describes how bed-form changes affect highways in the river environment. 1.

1. buildings. and rivers that are essentially straight. All material in a streambed will erode with time. are reasonably predictable in behavior. earth-filled.Scour at Bridges Scour at highway structures is the result of the erosive action of flowing water removing bed material from around the abutments and piers which support the bridge and bed and bank material of the stream the structure crosses. Recommended structures and design methods for river control are presented in Chapter 6. Meandering rivers that are not subject to rapid movement. meandering rivers are generally unstable with eroding banks which may result in destruction of productive land. Example problems related to riprap are solved at the end of Chapter 6 in SI and English units of measurement. The integrated and interactive effects of these structures with the river are discussed in Chapter 9. Bank protection works are often necessary to stabilize reaches of many rivers and to improve them for other aspects of flood control and navigation. A simple river classification scheme is presented. however.7. Example problems to determine river classification and response are given at the end of Chapter 5. angled and sloped rock-filled. shales. However. and concrete tetrahedrons have all been used in the practice of training rivers and stabilizing river banks.19 .6 Chapter 7 . control works. willow and asphalt mattresses. nick points.River Stabilization and Bank Protection Numerous types of river control and bank stabilization devices have evolved through past experience. 2001). timber piles. 2001). bridges. sacked concrete and sand. meandering. geomorphic thresholds. However.4 Chapter 5 . In general. sheet piles. Sandstone. HEC-23 has detailed design guidelines for river stabilization and protection countermeasures. Terms such as fluvial cycles. 1.7. braided rivers are relatively steep and meandering rivers have more gentle slopes. Additional information on stream stability at highway structures can be found in HEC-20 (Lagasse et al. while sandbed streams may erode to the maximum depth of scour in hours. and other sedimentary bedrock materials can erode to the 1.River Morphology and River Response Rivers have different alignments and geometry. and urban properties during floods. braided rivers. Detailed guidelines for selection and design of stream instability and bridge scour countermeasures are presented in HEC-23 (Lagasse et al. Chapter 5 presents the fluvial geomorphology of rivers and methods to predict river response to external forces such as a bridge crossing or other natural or human induced changes. and head cuts are described. Both scour at highway structures and stream migration (instability) can cause a bridge failure. brick.5 Chapter 6 . The study of river morphology and river response in Chapter 5 makes it clear that both shortand long-term changes can be expected on river systems as a result of natural and human influences. and timber dikes.1. bridge approaches. steel jack and brush jetties. riprap grouted slope protection. some material such as granite may take hundred's of years to erode. There are meandering rivers.7. automobile bodies. A variety of methods to predict a river's response to change are given. Concrete. alluvial fans.

The erosion of cohesive and other cemented material is slower than sand bed material. Cohesive bed and bank material such as clays. sediment transport. analysis and design of river crossings and encroachments. but the effects of scour can be cumulative in both cohesive and non-cohesive materials. Whereas. In addition to identifying data needed. The principal factors to be considered in design are presented.20 . with clear-water scour the scour depths are limited by the critical velocity or critical shear stress of a dominant size in the bed material at the crossing. It might take the erosive action of several major floods. 1. is necessary for a meaningful analysis of complex river response problems. will erode. silts and silty sands or even coarser bed material such as glacial tills. which are cemented by chemical action or compression. Chapter 9 contains a series of conceptual examples (cases) of river environments and their response to crossings and encroachments based upon geomorphic principles given in Chapter 5.7. and river mechanics to the hydraulic and environmental design of river crossings and highway encroachments. Transport of appreciable bed material into the crossing results in live-bed scour. Scour at bridge crossings is a sediment transport process. The design of most complex problems in river engineering can be facilitated by a qualitative evaluation combined with a quantitative analyses. In this latter case the transport of the bed material limits the scour depth. general scour.8 Chapter 9 .7 Chapter 8 – Data Need and Data Sources The purpose of Chapter 8 is to identify data needed for calculations and analyses for highway crossings and encroachments of rivers. followed by a quantitative estimate. a more detailed analysis of scour at highway bridges and detailed example problems in SI and English systems of measurement are presented in HEC-18 (Richardson and Davis 2001). However. followed by a discussion of the procedures recommended for the evaluation. 1. but their ultimate scour will be as deep if not deeper than the scour depth in a non-cohesive sandbed stream. fluvial geomorphology.7. silty clays. These histories document river response to highway crossings and encroachments and illustrate river response qualitatively. Chapter 7 presents the basic definitions of scour at bridges. If there is no transport of bed material into the bridge crossing.Design Considerations for Highway Encroachment and River Crossings Chapter 9 presents applications of the fundamentals of hydraulics. The hypothetical cases are followed by practical examples (actual case histories) for river crossings in the United States. The types and amounts of data needed for planning and designing river crossings and lateral encroachments depend upon the class of the proposed highway. These cases indicate the trend of change in river morphology for given initial conditions.extent that a bridge will be in danger unless the substructures are founded deep enough. Chapter 8 identifies sources of data. the systematic approach of a qualitative assessment of channel response. 1. and develops the basic equations for determining the scour components. hydrology. and local scour at piers and abutments result when more sediment is removed from these areas than is transported into them. Long-term degradation. In most cases. clear-water scour exists.

s). A three-level design procedure is emphasized in these designs. and river mechanics principles described in this manual to design safe and economical crossings that protect.31 ft /s 2 2 Acceleration of Gravity 9. This edition uses for the unit of length the meter (m) or foot (ft).32) °F 1.31 ft 3 3 Discharge 1 m /s 35. The designs use the geomorphic.7. but in calculations the decimal equivalent of millimeters in meters is used (1 mm = 0. 1. N/m ) or (lb/ft ). the designs are determined by wellestablished numerical procedures.2 ft/s 3 3 Unit Weight of Water 9800 N/m 62.1. Three examples are given to illustrate the application of the principles. Term SI Units English Units Length 1m 3. In the examples.11 Appendices The appendices present a discussion of the metric (SI) system (Appendix A).8 DUAL SYSTEM OF UNITS This manual uses dual units (SI metric and English).7.21 .94 slugs/ft 3 3 Density of Quartz 2647 kg/m 5. of weight/force the newton (N) or pound (lb). and of temperature the degree centigrade (°C) or Fahrenheit (°F).14 slugs/ft Specific Gravity of Quartz 2. The conversion factors. 1. The unit of time is the same in SI as in English system (seconds.4 lb/ft 3 3 Density of Water 1000 kg/m 1. Commonly Used Engineering Terms in SI and English Units.Overview Examples of Design for Highways in the River Environment The objective of Chapter 10 is to present overview examples of the use of the principles presented in this manual in the design of highway encroachments and crossings in the river environment. however. maintain and restore the river environment. the metric (SI) unit of measurement is explained. In Appendix A. sediment particle size grade scale. hydrologic.28 ft 3 3 Volume 1m 35.65 2. methods. Sediment particle size is given in millimeters (mm).References Chapter 11 contains an alphabetical listing of the references cited throughout the document.9 Chapter 10 . physical properties of water in SI system of units. of pressure the Pascal 2 2 (Pa. 1. they also depend heavily on the judgment of the engineer.1. of mass the kilogram (kg) or slug. analysis of additional sediment transport relationships (Appendix B).1. hydraulic.7.10 Chapter 11 . and concepts given in this manual. and an Index (Appendix C). and some common equivalent hydraulic units are also given. Table 1.81 m/s 32.001 m) or for the English system feet (ft). The value of some hydraulic engineering terms used in the text in SI units and their equivalent English units are given in Table 1.65 Specific Gravity of Water 1 1 Temperature °C = 5/9 (°F .

(page intentionally left blank) 1.22 .

1 . That is. steady or unsteady. and the two dimensionless numbers (the Froude number and Reynolds number) are more fully explained in the following sections. the flow is called critical. no deformation of the bed and banks is considered. if Fr > 1. 2. The mathematical representation of the fluid velocity is a function of the increment of length ds during the infinitesimal time dt. v= ds dt (2. it has magnitude and direction. Mobile boundary hydraulics will be discussed in later chapters. surface configuration. (3) laminar or turbulent flow. Open channel flow can be classified as: (1) uniform or nonuniform flow. laminar or turbulent. Acceleration: Acceleration is the time rate of change in magnitude or direction of the velocity vector. and (4) tranquil or rapid flow. In laminar flow. (2) steady or unsteady flow. In this chapter. the water surface is not confined. Mathematically. In steady flow. In uniform flow. Turbulent flow on the other hand is characterized by random fluid motion.1 Definitions Velocity: The velocity of a fluid particle is the time rate of displacement of the particle from one point to another. and if Fr = 1. the flow is supercritical. Because the classifying characteristics are independent. the velocity at a given depth does not change. Also. turbulent and rapid at the same time. Open channel flow can be nonuniform. If Fr < 1. In open channel flow. the depth and discharge remain constant with respect to space.1. the discussion is restricted to a one-dimensional analysis of rigid boundary open channel flow where velocity and acceleration are large only in one direction and are so small as to be negligible in all other directions.1) Streamline: An imaginary line within the flow which is everywhere tangent to the velocity vector is called a streamline. no change occurs with respect to time at a given point. the fundamentals of rigid boundary open channel flow are described.CHAPTER 2 OPEN CHANNEL FLOW 2.2) dt 2. the flow is subcritical. Tranquil flow is distinguished from rapid flow by a dimensionless number called the Froude number. one layer not mixing with adjacent ones. rapid or tranquil.1 INTRODUCTION In this chapter. acceleration a is expressed by the total derivative of the velocity vector or dv a= (2. Fr. In rigid boundary open channel flow. uniform or nonuniform. sixteen different types of flow can occur. the flow field can be characterized by layers of fluid. Velocity is a vector quantity. thus. Mobile boundary hydraulics refers to flow which can generate deformation of the boundary through scour and fill. These terms. flow pattern and pressure distribution within the flow depend on gravity. unsteady.

and surges. flood hydrographs. ∂v s v2 ≠ 0 and ≠0 ∂s r (2. the convective acceleration terms are not equal to zero. both magnitude and direction.2 . ∂v 2 v2 s = 0 and =0 ∂s r (2. This is called convective acceleration. both magnitude and direction.8) (2. Turbulent flow: In turbulent flow the mixing of the fluid and momentum transfer is related to random velocity fluctuations.6) æ ∂v ö ≠ 0 ÷ are examples of Flow around a bend ( v 2 / r ≠ 0) and flow in expansions or contractions ç ∂ s è ø nonuniform flow. with distance. the velocity at a point does not change with time ∂v s ∂v = 0 and n = 0 ∂t ∂t Unsteady flow: In unsteady flow. The second term in each equation is the change in velocity.4) The first terms in Equations 2. which is a dimensionless ratio of the inertial forces to the viscous forces.5) Nonuniform flow: In nonuniform flow. Laminar flow: In laminar flow. The flow is laminar or turbulent depending on the value of the Reynolds number (Re = ρVL/µ). V is the fluid velocity.7) Examples of unsteady flow are channel flows with waves. the tangential component embodying the change in magnitude of the velocity. and 2. and the normal component reflecting the change in direction dv s ∂v s 1 ∂ (v 2 s ) as = = + dt ∂t 2 ∂s (2. Steady flow: In steady flow.4 represent the change in velocity. with time at a given point. Uniform flow: In uniform flow the convective acceleration terms are zero. a. Here ρ and µ are the density and dynamic viscosity of the fluid.3 and 2.3) an = dv n ∂v n v 2 + = dt r ∂t (2.The vector acceleration. the velocity at a point varies with time ∂v s ∂v ≠ 0 and n ≠ 0 ∂t ∂t (2. has components both tangential and normal to the streamline. the mixing of the fluid and momentum transfer is by molecular activity. This is called the local acceleration.

The boundary condition that controls the tranquil flow depth is always located at the downstream end of the subcritical reach. The concepts explained in this chapter assume one-dimensional flow and the derivations of the equations utilize a control volume. the flow is supercritical (or rapid) and surface disturbances can propagate only in the downstream direction. viscous forces are dominant and Re is relatively small. (2) the conservation of linear momentum. designated section 2. The Froude number is also the ratio of the flow velocity V to the celerity (c = gL ) of a small gravity wave in the flow (this concept is detailed in Section 2. Re is large. inertial forces are very much greater than viscous forces. Rapid flow: When Fr > 1. 2. the flow is critical and surface disturbances remain stationary in the flow.0. the downstream cross-section. in response to changes in channel geometry depends on the Froude number (Fr = V / gL ) . the accelerations in other directions being negligible.L is a characteristic dimension. The control section of rapid flow depth is always at the upstream end of the rapid flow region. the flow is subcritical (or tranquil). Tranquil flow: In open channel flow. momentum. Turbulent flows are predominant in nature. the free surface configuration. The boundaries of the control volume are the upstream cross-section. In turbulent flow. 2. through which mass. and the interface between the water and the wetted perimeter (banks and bed). usually the depth (or the hydraulic radius) in open channel flow.5).1 Conservation of Mass Consider a short reach of river shown in Figure 2. The conservation of mass is another way of stating that (except for mass-energy interchange) matter can neither be created nor destroyed.2 THREE BASIC EQUATIONS The basic equations of flow in open channels are derived from the three conservation laws. the free surface of the water between sections 1 and 2. Laminar flow occurs very infrequently in open channel flow. The control volume may be assumed fixed in space or moving with the fluid. The principle of conservation of linear momentum is based on Newton's second law of motion which states that a mass (of fluid) accelerates in the direction of and in proportion to the applied forces on the mass. However. designated section 1. A control volume is an isolated volume in the body of the fluid.3 . In the analysis of flow problems.2. When Fr < 1. which is the ratio of inertial forces to gravitational forces. When Fr = 1.1 as a control volume. and (3) the conservation of energy. These are: (1) the conservation of mass. a very inaccurate analysis may occur if one assumes accelerations are small or zero when in fact they are not. and energy can be convected. 2. and surface waves propagate upstream as well as downstream. that is. In laminar flow. much simplification can result if there is no acceleration of the flow or if the acceleration is primarily in one direction.

The statement of the conservation of mass for this control volume is Mass flux out of the control volume Mass flux into the control volume + Time rate of change in mass in the control volume =0 Mass can enter or leave the control volume through any or all of the control volume surfaces. and may be written as: Mass flux out of the control volume Similarly Mass flux into the control volume The amount of mass inside a differential volume dV inside the control volume is ρ dV and = = ò A ρ 2 v 2 dA 2 2 (2. the mass flux out of the control volume through the differential area dA2 is ρ2 v2 dA2.4 . In the absence of rainfall. seepage and other lateral mass fluxes.9) ò A ρ1 v 1 dA 1 1 (2. At section 2.Figure 2. evaporation. A river reach as a control volume.1. The total mass flux out of the control volume at section 2 is the integral of all ρ2 v2 dA2 through the differential areas that make up the cross-section area A2. mass enters the control volume at section 1 and leaves at section 2. Rainfall would contribute mass through the surface of the control volume and seepage passes through the interface between the water and the banks and bed. The values of ρ2 and v2 can vary from position to position across the width and throughout the depth of flow at section 2.10) 2.

Equation 2.14) A The symbol v represents the local velocity whereas the velocity V is the average velocity at the cross-section.13 reduces to the statement that inflow equals outflow or ρV2 A 2 − ρV1 A 1 = 0 That is. (2. the flow is steady and there is no significant lateral inflow or seepage. ρ1 = ρ2 = ρ. so we define an average velocity V such that V= 1 ò A v dA (2.15) and Equation 2. It is applicable when the fluid density is constant.16 is the familiar form of the conservation of mass equation for steady flow in rivers.13) It is often convenient to work with average conditions at a cross-section.16) 2. for steady flow of incompressible fluids V1 A 1 = V2 A 2 = Q = VA where Q is the volume flow rate or the discharge. In mathematical notation.11) The statement of conservation of mass for the control volume calls for the time rate of change in mass. Because water is nearly incompressible the density ρ of the fluid is considered constant.12) For the reach of river. When the flow is steady ∂ ∂t ∀ ò ρ d∀ = 0 (2. the statement of the conservation of mass becomes ò A ρ 2 v 2 dA 2 − ò A ρ1 v 1 dA 1 + ∂t ò ∀ ρ d∀ = 0 2 1 ∂ (2.Mass inside the = control volume ò ∀ ρ d∀ (2. Time rate of change in mass in the control volume = ∂ ò ∀ ρ d∀ ∂t (2.5 .

Therefore.18) 2. The statement of conservation of linear momentum is: Flux of momentum Flux of Momentum . as a starting point.2.2). the pressure terms p1 and p2 are directed toward the control volume in a direction normal to the Sections 1 and 2. the flux of momentum out of the control volume through the differential area dA2 is: ρ2v2 dA2 v2 (2. = ò A ρ 2 v 2 dA 2 v 2 2 (2.17) Here ρ2 v2 dA2 is the mass flux (mass per unit of time) and ρ2 v2 dA2 v2 is the momentum flux through the area dA2.6 .2.1 is rather complex to analyze in terms of Newton's Second Law because of the curvature in the flow. at the inflow section (section 1). Consider the conservation of momentum in the direction of flow (the x-direction in Figure 2. Flux of momentum out of the control volume Similarly. For this control volume. The terms in the statement are vectors so we must be concerned with direction as well as magnitude.into the control out of the control volume volume Time rate of change = + of momentum in the control volume Sum of the forces acting on the fluid in the control volume Figure 2.2. At the outflow section (section 2). the differential length of reach dx is isolated as a control volume.2 Conservation of Linear Momentum The curved reach of the river shown in Figure 2. shown in Figure 2. The shear stress τo is exerted along the interface between the water and the wetted perimeter and is acting in a direction opposite to the axis x. The control volume for conservation of linear momentum.2.

There is a fluid shear stress τo acting along the interface between the water and the bed and banks.22 to give ρβ 2 V22 A 2 − ρβ1 V12 A 1 = ò A1 p 1 dA 1 − ò A2 p 2 dA 2 − ò L τ o P dx + Fb (2.22) For steady incompressible flow.Flux of momentum into the control volume = ò A ρ1 v 1 dA 1 v 1 1 (2. as with the conservation of mass equation. The total force in the x-direction at section 1 is ò A1 p 1 dA 1 . it is convenient to use average velocities instead of point velocities.7 . Equation 2.19) The amount of momentum in the control volume is ò ∀ ρv dV so Time rate of change of momentum = in the control volume ∂ { ò ∀ ρv dV } ∂t (2. The momentum coefficient for incompressible fluids is: β= 2 ò A v dA V A 2 1 (2. the force acting on the differential area dA1 of the control volume is p1 dA1 where p1 is the pressure from the upstream fluid on the differential area.23) The pressure force and shear force terms on the right-hand side of Equation 2. Similarly. The shear on the control volume is in a direction opposite to the direction of flow and results in a force -τoP dx where τo is the average shear stress on the interface area. The term P dx is the interface area.21 is combined with Equation 2.21) Again.24) 2. The statement of conservation of momentum in the x-direction for the control volume is: 2 2 ò A ρ 2 v 2 dA 2 − ò A ρ1 v 1 dA 1 + ∂t ò ∀ ρv dV = ò A p 1 dA 1 − ò A p 2 dA 2 − ò L τ o P dx + Fb 2 1 1 2 ∂ (2. at section 2. P is the average wetted perimeter and dx is the length of the control volume.23 are usually abbreviated as Σ Fx so: å Fx = ò A1 p1 dA 1 − ò A2 p 2 dA 2 − ò L τ o P dx + Fb (2. the total force is ò A 2 p 2 dA 2 .20) At the upstream section. We define a momentum coefficient β so that when average velocities are used instead of point velocities. The body force component acting in the x-direction is denoted Fb and will be discussed in a subsequent section. the correct momentum flux is considered.

The streamtube as a control volume. 2.26) The First Law of Thermodynamics can be written: Q−W = where: Q • • • dE dt (2.8 . Figure 2.25) for steady flow with constant density.The conservation of momentum equation becomes: ρβ 2 V22 A 2 − ρβ1 V12 A 1 = å Fx (2. the control volume is reduced to the size of a streamtube connecting dA1 and dA2 as shown in Figure 2. With Equation 2.3 Conservation of Energy (2.3. The statement of conservation of energy for a control volume is then: Flux of energy out of the control volume Flux of energy into the control volume + Time rate of change of energy in the control volume = Q−W • • The choice of a control volume is arbitrary. The streamtube is bounded by streamlines through which no mass or momentum enters.3.2.16 the steady flow conservation of linear momentum equation takes on the familiar form ρQ (β 2 V2 − β1 V1 ) = å Fx 2.27) = = = Rate at which heat is added to a fluid system Rate at which a fluid system does work on its surroundings Energy of the system W E • Then dE/dt is the rate of change of energy in the system. To illustrate the procedure.

The rate at which the fluid pressure does work on the control volume surrounding through the boundary dA1 in Figure 2.28) Similarly.32) The work done by the fluid in the control volume on its surroundings can be in the form of pressure work W p. Hence. for the streamtube W p = p 2 dA 2 v 2 − p 1 dA 1 v 1 • (2. the rate at which heat is added to the control volume can be neglected.30) = ρ 1 e1 dA 1 v 1 (2. evaporation losses. Flux of energy into the control volume and Time rate of change of energy = in the control volume 0 (2.33) 2. no shaft work is involved (W s = 0). the rate of doing pressure work is p 2 dA 2 v 2 At the other boundaries of the streamtube.29) Here e is the energy per unit mass. or problems concerning the formation of ice in rivers.9 .For steady flow of an incompressible fluid in the streamtube Flux of energy out of the control volume = ρ 2 e 2 dA 2 v 2 (2.31) Unless one is concerned with thermal pollution. that is: Q~ −0 • (2.3. or shaft work (mechanical work) W s. For the streamtube shown in Figure 2. Accordingly.3 is: − p 1 dA 1 v 1 and on boundary dA2. at the inflow section (section 1). shear work W τ. there is no pressure work because there is no fluid motion normal to the boundary. the total energy E in a control volume is: E = ò ∀ ρe d∀ (2.

10 .3 is: ρ 2 e 2 v 2 dA 2 − ρ 1 e1 v 1 dA 1 = p 1 v 1 dA 1 − p 2 v 2 dA 2 − τPv dx (2.35) The conservation of mass for steady flow in the streamtube is (according to Equation 2.36) This expression for e is substituted in Equation 2.37) (2.16) v 2 dA 2 = v 1 dA 1 = dQ Now Equation 2. kinetic and potential energies or e =u+ where: u v g z = = = = Internal energy associated with the fluid temperature Velocity of the mass fluid Acceleration due to gravity Elevation above some arbitrary reference level v2 + gz 2 (2. The rate at which the fluid in the streamtube does shear work on the control volume is: W τ = τP dx v • (2.35 reduces to (ρ 1 e 1 + p 1 ) dQ − (ρ 2 e 2 + p 2 ) dQ = τPv dx The energy per unit mass e is the sum of the internal. P is the average perimeter of the streamtube.Along the interior boundaries of the streamtube there is a shear stress resulting from the condition that the fluid velocity inside the streamtube may not be the same as the velocity of the fluid surrounding the streamtube.38) (2. dx is the length of the streamtube and v is the fluid velocity at the streamtube boundary.40) 2.39) By dividing through by g and defining the head loss h l as follows: hl = u 2 − u1 τPvdx + g ρgdQ (2. Then for steady flow in the streamtube.34) where τ is the average shear stress on the streamtube boundary. the statement of the conservation of energy in the streamtube shown in Figure 2.37 to yield u1 + 2 p v1 p v2 τPvdx + gz 1 1 = u 2 + 2 + gz 2 + 2 + 2 ρ 2 ρ ρdQ (2. The product P dx is the surface of the streamtube subjected to shear stresses.

there is not sufficient information available to do a differential streamtube analysis of a reach of river.41) If there is no shear stress on the streamtube boundary and if there is no change in internal energy (u1 = u2).41 which can be written: 2 æ v1 ö æ v2 ö p1 p2 2 ç ÷ ç + + z v dA = 1 ç 2g γ ÷ ç 2g + γ + z 2 ÷ ÷ v dA + h l vdA è ø è ø (2.43) because v1 dA1 = v2 dA2 = vdA for the streamtube.43 over the cross-section area: 2 ì ü æ v2 p ö p ï αV ï ç ÷ + + z vdA = + + zý Q í ò A ç 2g γ ÷ γ ï ï è ø î 2g þ (2.42) which is the Bernoulli Equation.46) A 2. We know the statement of the conservation of energy for a streamtube.45) v 3 dA to allow the use of average velocity V rather than point velocity v. A reach of river such as that shown in Figure 2.44) where: α α= 1 ò V3A A = Kinetic energy correction factor defined by the expression (2.11 . The average pressure over the cross-section is p . so appropriate changes must be made in the energy equation.The energy equation for the streamtube becomes 2 v1 p v2 p + 1 + z1 = 2 + 2 + z 2 + h l 2g γ 2g γ (2. the energy equation reduces to: 2 v1 p v2 p + 1 + z 1 = 2 + 2 + z 2 = cons tan t 2g γ 2g γ (2. defined as: p= 1 ò VA p vdA (2. It is Equation 2. The common form of the energy equation used in open channel flow is derived by integrating Equation 2. Generally.1 can be pictured as a bundle of streamtubes.

12 . Usually it is assumed that the pressure is hydrostatic and the average elevation head z is at the centroid of the cross-sectional area.45.50) The tendency in river work is to neglect the energy correction factor even though its value may be as large as 1.46.5. the piezometric head varies in the flow field.51) In steady uniform flow (and for zero flow).48) In summary. the differential equation of motion in an arbitrary direction x is ö ax ∂ æp ç + z÷ ç ÷= g ∂x è γ ø (2. 2. This is illustrated in Figure 2.4a the pressure at the bed is hydrostatic and equal to γyo whereas in the curvilinear flow (Figure 2. the expression for conservation of energy for steady flow in a reach of river is written α 1 V12 p 1 α V2 p + + z 1 = 2 2 + 2 + z 2 + HL 2g γ 2g γ (2.52) However. and 2.3 HYDROSTATICS When the only forces acting on the fluid are pressure and fluid weight. when there is acceleration.14 (SI) and 2. it should be kept in mind that Equations 2.4.47 are the correct definitions of the terms in the energy equation. the piezometric head is not constant in the flow.15 (English).4b) the pressure is larger than γyo because of the acceleration resulting from a change in direction.47) A and Q is the volume flow rate or the discharge. the acceleration is zero and we obtain the equation of hydrostatics p + z = Cons tan t γ (2.The term z is the average elevation of the cross-section defined by the expression: z= 1 ò VA zvdA (2. 2. 2. By definition Q=ò Also HL = 1 ò A h l vda VA (2. In Figure 2. However. An example problem illustrating the calculation of α and β for a stream is provided in Section 2. That is.49) A vdA (2.

13 . Figure 2. expanding or have substantial curvature (curvilinear flow).Figure 2. for rapidly varying flow where the streamlines are converging.53) Note that y is the depth (perpendicular to the water surface) to the point.5. the constant is equal to zero for gage pressure at the free surface of a liquid. For most channels. For flow with hydrostatic pressure throughout (steady. uniform flow or gradually varied flow).4. as shown in Figure 2. it follows that the pressure head p/γ is equal to the vertical distance below the free surface. In sloping channels with steady uniform flow. Pressure distribution in steady uniform flow on steep slopes (after Chow 1959). However. 2. when fluid acceleration is small (as in gradually varied flow) the pressure distribution is considered hydrostatic. the pressure head p/γ at a depth y below the surface is equal to p = y cos θ γ (2. Pressure distribution in steady uniform and in steady nonuniform flow. In Equation 2. θ is small and cosθ ~ 1.5. In general. fluid accelerations are not small and the pressure distribution is not hydrostatic.52.

streamlines are straight and parallel. − p1 ~ γy 2 cos θ 2 (2.1 Introduction In steady.55) Assuming only small variations in the point velocity v with y we have: − p1 ~ γy 1 cos θ 2 (2. Then according to Equation 2.57) 2. the conservation of energy for this control volume is: Figure 2.46 p1 = 1 y1 ò o γy cos θ v 1 dy V1 y 1 (2.50.54) The pressure at any point y below the surface is y cosθ. According to Equation 2. uniform open channel flow there are no accelerations.6. The slope of the water surface Sw and the bed surface So and the energy gradient Sf are equal. and the pressure distribution is hydrostatic.4.6 as a control volume.14 . Steady uniform flow in a unit width channel.4 STEADY UNIFORM FLOW 2. Consider the unit width of channel shown in Figure 2.2. α 1 V12 p 1 α V2 p + + z 1 = 2 2 + 2 + z 2 + HL 2g γ 2g γ (2.56) Similarly.

63) (2.65) Steady uniform flow is an idealized concept for open channel flow and is difficult to obtain even in laboratory flumes.58) With the above expressions for p 1. depth or direction (resulting in nonuniform flow) are so small or occur over such a long distance that the flow can be considered uniform. and z 2 the energy equation for this control volume reduces to 2 y 2 cos θ α 1 V12 y 1 cos θ y cos θ α 2 V2 y cos θ + + z1 + 1 = + 2 + z2 + + HL 2g 2 2 2g 2 2 (2. 2.62) (z1 + y 1 ) − (z 2 + y 2 ) ∆L (2. p 2 . water surface and energy grade line are respectively S o = sin θ = Sw = and (z1 − z 2 ) ∆L (2.47 z1 ~ − z1 + and − z2 + z2 ~ y 2 cos θ 2 (2.60) or α 1 V12 α V2 + y 1 cos θ + z 1 = 2 2 + y 2 cos θ + z 2 + HL 2g 2g (2.15 . z 1.59) y 1 cos θ 2 (2. the flow is steady and the changes in width. The velocity distribution in the vertical is normally a log function for which α1 _ α2 _ 1. Then the energy equation becomes: V12 V2 + y 1 + z 1 = 2 + y 2 + z 2 + HL 2g 2g and the slopes of the bed. For many applications.Also according to Equation 2.61) For most natural channels θ is small and y cosθ _ y.64) æ V12 ö æ V22 ö ç ÷ ç ÷ + y + z − + y + z 1 1 2 2 ÷ ç 2g ÷ 2g HL ç è ø è ø Sf = = ∆L ∆L (2.

if the roughness at the boundary is submerged in a viscous sublayer as shown in Figure 2.66) The velocity gradient is evaluated at the boundary. These variables are interrelated. The random fluctuation in velocity characterizes turbulent flows.Variables of interest for steady uniform flow are: (1) the mean velocity V. Velocity Distribution. In laminar flow.7. 2.2 Shear Stress. the particle has a vertical component of velocity vy as well as a horizontal component vx. The paths of fluid particles in the vicinity of the boundary are shown in Figure 2. (4) the head loss HL through the reach. The shear stress exists only when fluids are in motion. When the boundary is hydraulically rough. and Average Velocity Shear stress τ is the internal fluid stress which resists deformation. which is a normal stress. that is. As shown in Figure 2. 2. the shear stress at the boundary is: æ dv ö τo = µ ç ç dy ÷ ÷ at y = 0 è ø (2. Figure 2. The dynamic viscosity µ is the proportionality constant relating boundary shear and velocity gradient in the viscous sublayer.8. The local shear stress at the interface between the boundary and the fluid can be determined quite easily if the boundary is hydraulically smooth. the thickness of the laminar sublayer is very small compared to the roughness height.9a. It is a tangential stress in contrast to pressure. Here. and (5) the shear stress. the thickness of the laminar sublayer is denoted δ′.7.4.16 . The velocity at a point near the boundary fluctuates randomly about a mean value. both local τ and at the bed τo. Hydraulically smooth boundary. (3) the velocity distribution v(y) in the vertical. (2) the discharge Q.

67) where v x and v y are the time-averaged mean velocities in the x and y direction and vx′ and vy′ are the fluctuating components. Hydraulically rough boundary.68) (2.8.9.9a can be written as: v x = v x + v ′x and v y = v y + v ′y (2.17 . Velocities in turbulent flow. Through theoretical investigation it has been found that turbulence generates shear stress given by τ = − ρv ′x v ′y (2. The two components of velocity in Figure 2.69) 2. Figure 2.Figure 2.

For rigid boundaries κ has the average value of 0. v ′x ~ l dv dy dv dy (2.70) v ′y ~ l and (2.18 . The term τo is the bed shear stress. It is called the Reynolds shear stress.71) æ dv ö τ~ρl ç ç dy ÷ ÷ è ø 2 2 (2. then for steady uniform turbulent flow. V*. The term y′ depends on the flow and has been experimentally determined.74) where κ is the von Karman universal velocity coefficient.73 can be rearranged to the form: τo / ρ dv = dy κy (2. Integration of Equation 2.. The term (τo/ρ)1/2 has the dimensions of velocity and is called the shear velocity. æ dv ö τ = ρκ y ç ç dy ÷ ÷ è ø 2 2 2 (2.e. The term y′ results from evaluation of the constant of integration assuming v = 0 at some distance y′ above the bed.75) Here In is the logarithm to the base e and log is the logarithm to the base 10.9b.72) If it is assumed that the mixing length can be represented by the product of a constant κ and y (i. Accordingly.73) Using different reasoning von Karman (1930) derived the same equation.4. which is assumed to be the same in both the x and y directions. The many experiments have resulted in characterizing turbulent flow into three general types: 2.The term v ′x v ′y is the time-average of the product of v ′x and v ′y at a point in the flow. l = κy). Prandtl (1925) suggested that v ′x and v ′y are related to the velocity gradient dv/dy shown in Figure 2.74 yields v τo ρ = y y 2.31 1 log In = y′ κ y′ κ (2. He proposed to characterize the turbulence with a dimension called the "mixing length" l. Equation 2.

27 ks ks V* g (2.76) δ′ = (2) Hydraulically rough boundary turbulent flow where velocity distribution.75 log 12. Then with 11. the velocity distribution v.(1) Hydraulically smooth boundary turbulent flow where the velocity distribution.75 log ç 30. æ Xy ö v ÷ = 2.10 Height of the roughness elements.2 k ÷ ÷ s ø è (2.27 o = 2.2 ç V* ks ÷ è ø and æ Xy ö ç ç 30.2 where ks is the height of the roughness element. mean velocity and resistance to flow depend on both fluid viscosity and boundary roughness. Transition where the velocity distribution. and resistance to flow equations can be written in the following dimensionless form which is related to the above flow types by Figure 2.78 and 2. The symbols of Equations 2.80) 2. Then (2.10.79 denote: X ks v yo V V* τo R = = = = = = = = Coefficient given in Figure 2.6 υ τo ρ we find y ′ = δ′ 107 (2.10. equal to the cross-sectional area A divided by the wetted perimeter P (2.5 In 12. mean velocity and resistance to flow are independent of the boundary roughness of the bed but depend on fluid kinematic viscosity.79) Note that any system of units can be used as long as yo and ks (and V. mean velocity and resistance to flow are independent of viscosity and depend entirely on the boundary roughness. for sand channels Local mean velocity at depth y Depth of flow Depth-averaged velocity Shear velocity τ o / ρ which for steady uniform flow is gRS f Shear stress at the boundary and for steady uniform flow the average is τ o = γRS f Hydraulic radius.77) (3) k δ′ < y′ < s 107 30.19 . As a result.2 The boundary roughness effects can be merged into one equation by using y′ = ks/(30. mean velocity V. v and V*) have the same dimensions.2X) where X is determined from Figure 2.78) Xy Xy o V C = = 5. y′ = ks/30.5 In = 5. For this case.

6ν / V* C / g = Chezy discharge coefficient in the equation æ 8g ö V = C RS or C = ç ÷ è f ø f f =8 τo ρV 2 = 1/ 2 (2. Einstein's multiplication factor X in the logarithmic velocity equations (Einstein 1950).10.81) Darcy-Weisbach resistance coefficient which is given by the expression (2.82) Any consistent set of dimensions can be used (m or ft).20 . 2.Figure 2. Sf δ′ ν = = = Slope of the energy gradeline Thickness of the viscous sublayer Kinematic viscosity of fluid δ ′ = 11.

Two such equations are in common use. 2. other approaches to determine mean velocities in rivers has been prevalent.4. By equating the velocity determined from Manning's equation with the velocity determined from Chezy's equation. They are Manning's equation: V= Ku /2 R 2 / 3 S1 f n (2.83) and Chezy's equation /2 V = CR 1/ 2 S 1 f (2.1. ft/s Manning's roughness coefficient Hydraulic radius in m or ft equal to the cross-sectional area A divided by the wetted perimeter P of the waterway m. the relation between the coefficients is C= K u 1/ 6 R n (2.2. ft Friction slope m/m.85) If the flow is gradually varied. ft/ft Chezy's discharge coefficient known as Chezy's C 1. Additional values are given by Barnes (1967) and Chow (1959).84) where: V n R Sf C Ku Ku = = = = = = = Average velocity in the waterway cross-section in m/s.0 (SI) 1. a catalog of values of Manning's n and Chezy's C has been assembled so that an engineer can estimate the appropriate value by knowing the general nature of the channel boundaries. or both. the boundary shear stress is expressed implicitly in the roughness coefficient "n" or in the discharge coefficient C.21 . The term S fave is determined by averaging over a short time increment at a station or over a short length increment 300 m (1. Over many decades.486 (English) In these equations. An abbreviated list of Manning's roughness coefficients is given in Table 2. Manning's and Chezy's equations are used with the average friction slope S fave .3 Other Velocity Equations Because of the difficulties involved in determining the shear stress and hence the velocity distribution in turbulent flows. Manning's n for sandbed and gravel-bed channels is discussed in detail in Chapter 3.000 ft ) for example at an instant of time.

Clean. deep pools.0. No Brush 1.045 0. Bottom: cobbles with large boulders Floodplains 0.045-0. No Crop 2.Table 2.050 0.1.040 0. Mature Row Crops 3. more ineffective slopes and sections Same as 4.020-0. 6.014 0. lower stages.017 0.055 0. Mature Field Crops 0. no rifts or deep pools Same as above.035-0. cobbles.025-0. Short Grass 2. High Grass Cultivated Areas 1.030-0.045 0.050-0.040. 7. but some weeds and stones Same as above. but more stones Sluggish reaches.040 0. winding. Bottom: gravels. no vegetation in channel.015 0.070 Pasture.060 0. some pools and shoals Same as above.050 0. 5.030-0.012 0. but more stones and weeds Clean.22 . and earth channels in best condition Straight unlined earth canals in good condition Mountain streams with rocky beds Minor Streams (top width at flood stage < 30 m (100 ft) 0. trees and brush along banks submerged at high stages 1. Manning's n Rigid Boundary Channels Very smooth concrete and planed timber Smooth concrete Ordinary concrete lining Wood Vitrified clay Shot concrete.013 0.033 0. untrowelled. Manning's Roughness Coefficients for Various Boundaries.030-0. 3.080 0.025-0.0.050 2.030-0.025-0.020 0.075-0.150 Mountain Streams.050 0. 4. full stage.035 0.040-0. straight. or floodways with heavy stand of timber and underbush 0. deep pools Very weedy reaches.011 0. 2. 8.040 . weedy. banks usually steep.033-0. and few boulders 2.050 Streams on Plain 1.

050 0.160 Trees 1.050-0. 2.020 0. flood stage below branches 5.014 . Plane bed Ripples Dunes Washed out dunes or transition Plane bed 0.0.040 0.0.012 . 2. Scattered brush. 3. 4.200 0. Fr > 1 1. 5. 5.23 .018 . little undergrowth. but with flood stage reaching branches Major Streams (Top width at flood stage > 30 m (100 ft) The n value is less than that for minor streams of similar description.010 .080-0.025-0.120 0. Manning's n Floodplains (continued) Brush 1.035-0.030 0. because banks offer less effective resistance.035-0. Antidunes 1 0.0.060 0.060 0.0.013 Rapid Flow. 3. a few down trees. straight Cleared land with tree stumps.Table 2. 3.015 0.0. summer.110-0.020 .0 mm. 2.070-0. Fr < 1 1.080 0.020 Data is limited to sand channels with D50 < 1. Manning's Roughness Coefficients for Various Boundaries. 0.025 0.160 Regular section with no boulders or brush Irregular and rough section Alluvial Sandbed Channels (no vegetation)1 0. 2.035-0. no sprouts Same as above. Same as above.1.030-0. details to be discussed in Chapter 3.100 Tranquil flow. Standing waves 2.0.0.014 .110 0.080 0.010 .045-0.070 0. heavy weeds Light brush and trees in winter Light brush and trees in summer Medium to dense brush in winter Medium to dense brush in summer 0. 4. but with heavy growth of sprouts Heavy stand of timber.100-0.040-0. 4. Dense willows.

11. For example. and when the ice breaks up. For steeper streams. smooth channel in the materials involved. the ice layer may freeze into bank stabilization materials. resistance to flow is large on the floodplains. the reader is also referred to the work of Jarrett (1984. loading ramps. large quantities of rock and other material embedded in the ice may be floated downstream and subsequently thawed loose and dumped randomly leaving banks raw and unprotected.16 is then: 2. This surface shear is much larger than the normal shear stresses developed at the air-water interface. Resistance factors reflecting these effects must be selected largely on the basis of past experience with similar conditions. in and near water structures such as docks.24 .The general approach for estimating n values consists of the selection of a base roughness value for a straight. and ships. 1985). varying slopes and other complexities. buildings. decreasing the average velocity and increasing the depth. uniform. Benson and Dalrymple (1967) and Aldridge and Garrett (1973). Typical values are given in Table 2. Another serious effect is its influence on bank stability. undefined direction of flow. In some instances. natural and artificial irregularities. Chow (1959). the underside of the ice is deformed so that it resembles ripples or dunes observed on the bed of sandbed channels.86) Detailed values of the coefficients are found in Cowan (1956). The ice-water interface is not always smooth.2. When an ice cover occurs. then additive values are considered for the channel under consideration: n = (n o + n1 + n 2 + n 3 + n 4 ) m 5 where: no n1 n2 n3 n4 m5 = = = = = = Base value for straight uniform channels Additive value due to cross-section irregularity Additive value due to variations of the channel Additive value due to obstructions Additive value due to vegetation Mulitiplication factor due to sinuosity (2. 2. In general. There is an added shear stress developed between the flowing water and the ice cover. The conservation of mass Equation 2. the drag of ice retards flow.4 Average Boundary Shear Stress The average shear stress at the boundary τo for steady uniform flow is determined by applying the conservation of mass and momentum principles to the control volume shown in Figure 2. conditions are further complicated by deposition of sediment and development of dunes and bars which affect resistance to flow and direction of flow.4. With total or partial ice cover. The presence of ice affects channel roughness and resistance to flow in various ways. the open channel is more nearly comparable to a closed conduit. In many instances. Arcement and Schneider (1984) proposed a guide for selecting Manning's roughness coefficients for floodplains. The roughness characteristics on the floodplain are complicated by the presence of vegetation. This may cause overall resistance to flow in the channel to be further increased.

025-0.001-0.5 Obstructions 2.015 0.060 0.010-0.030 0.010 0.020-0.00 1.001-0. Adjustment Factors for the Determination of n Values.006-0.2.015 0-0004 0.010 0.25 .002-0.5 Vegetation Height Sinuosity < 1.025 0.2 < Sinuosity < 1. n1 Cross-Section Irregularity Conditions Smooth Minor Moderate Severe n2 Variations In Channel Section Gradual Alternating Occasionally Alternating Frequently n3 Negligible Minor Moderate Severe n4 Vegetation Small Medium Large Very Large m5 Sinuosity Minor Moderate Severe n Value 0 0.005 0.005 0.020 0 0.050-0.100 1.011-0.Table 2.30 Remarks Smoothest Channel Slightly Eroded Side Slopes Moderately Rough Bed and Banks Badly Sloughed & Scalloped Banks Gradual Changes Occasional Shifts From Large to Small Sections Frequent Changes in Cross-Sectional Shape Obstructions < 5% of Cross-Section Area Obstructions < 15% of Cross-Section Area Obstructions 15-50% of Cross-Section Area Obstructions > 50% of Cross-Section Area Flow Depth > 2 x Vegetation Height Flow Depth > Vegetation Height Flow Depth < Vegetation Height Flow Depth < 0.5 Sinuosity > 1.005-0.2 1.010-0.15 1.040-0.050 0.

91) With the above expressions for the components.11.Figure 2. the statement of conservation of linear momentum becomes: ρβ Wy o V 2 − ρβ Wy o V 2 = γAL sin θ + which reduces to τo = γ A sin θ P (2.25 with A1 = A2 = Wyo and V1 = V2.89) The downstream component of the body force γAL (equal to the weight of fluid in the control volume) in the X direction is: Fb = γAL sin θ (2.92) 2.87) The conservation of momentum in the downstream direction is described from Equation 2. The pressure forces acting on the control boundary are approximated by: F1 = F2 = 2 γ yo W 2 (2. ρy o WV 2 − ρy o WV1 = 0 or V1 = V2 (2. The shear force Fs in the x-direction is: Fs = τ o PL (2.26 . Control volume for steady uniform flow.93) 2 2 W γy o W γy o − − τ o PL 2 2 (2. The average boundary shear stress is τo acting on the wetted perimeter P.88) (2.90) where θ is the slope angle of the channel bed.

4. sin θ ~ − So and for steady uniform flow the average shear stress on the boundary is τ o = γRS o If the flow is gradually varied nonuniform flow. the local and average shear stress on the bed can be estimated by employing the velocity profile equations in Section 2. The equation is: 2.99) If the depth of flow yo.78 the local shear stress can be determined.75 log ç ç ÷ú k ê s è øú ë û 2 (2. However.97) For steady uniform flow. then the average shear stress can be determined from Equation 2. if two point velocities in a vertical profile are known (preferably in the lower 15 percent of the depth) the local shear stress on the bed can be determined from the following equation: τo = ρ( v 1 − v 2 ) 2 é æ y 1 öù ê5.98) This equation and the ones given below are valid for fully turbulent uniform flow over a hydraulically rough boundary in wide channels with a plane bed. 2.96) Measuring the average bottom shear stress directly in the field is tenuous. the average boundary shear stress is τ o = γRS f where Sf is the slope of the energy grade line.4. If the local velocity v1 at depth y1 is known and X = 1 for hydraulically rough boundary then.27 . If the channel slope angle is small. Alternatively. The equation is: τo = 2 ρv 1 é æ y 1 öù ÷ 30 . from Equation 2.5 Relation Between Shear Stress and Velocity (2.The term A/P is the hydraulic radius R.75 log ç ÷ú ç y ÷ ê ú è 2 øû ë 2 (2. the grain roughness ks and the average velocity in the vertical V are known.2.94) (2. 2 ê5.79.95) (2. the average bottom shear stress can be computed from the expression τ o = γRS f (2.

75 log ç ç12.6 Energy and Momentum Coefficients for Rivers In prismatic or constructed channels it is common to assume that the energy coefficient α and the momentum coefficient β are unity.4. the momentum coefficient β′ is (2. 5 í ç ç ÷ ÷ý V* ï î y o − y ′ è è y ′ ø øï þ For a vertical section of unit width.101) (2.103) the value of y for which Equation 2. From Equations 2.104) 2.22) The velocity distribution in wide channels for turbulent flow over a rough boundary is given by Equation 2. the depth of flow and the lower limit is y′ = ks 30.22: α= and β= 1 ò V2A A 1 ò V3A A v 3 dA (2. this is usually not the case.28 .78 with X = 1.27 k ê s è ë öù ÷ ÷ú ú øû 2 (2.5 V* 1 yo ~ ò 0 v dy − yo y o − y′ y ò y′ In ( y / y ′) dy o (2.100) 2.0 v = 2.τo = ρV 2 é æ yo ê5.102 yields ì æ y ö öü V ï yo æ ç In ç o ÷ − 1÷ï = 2.102) Here.78 gives a zero velocity. In river channels. the upper limit of integration is yo.45 and 2.45) v 2 dA (2. The integration of Equation 2.5 In (30.2 (2.2 y / k s ) V* The average velocity in the vertical is V= 2.

104 into Equation 2. 2. yo is the depth of flow at location w.β′ = y 2 dy ò y' v V ( y o − y ′) o 1 2 (2. The total discharge is Q and the total cross-sectional area is A.78 and 2. the energy coefficient for a vertical section unit width is 1 ò V 3 ( y o − y ′) α′ = yo y′ v 3 dy (2.45 can be written 1 æQö ç ÷ A èAø 3 α= [ò W 0 y 3 ò 0 v dy dw ] o (2. w is the lateral location of any vertical section.12.29 . so a graph of α′ and β′ vs yo/ks has been prepared. and v is the local velocity at the position y. 11 ç ks ÷ è ø 1 2 ì æ yo ö 2y ′ ü ïæ y o ö ï ç ÷ ÷ 2 + − íç In ÷ − 2 In ç ý ç ÷ y′ ø yo ï è y′ ø ï îè þ (2.105) If we substitute Equations 2.108 are rather complex. For the entire river cross-section (shown in Figure 2.107) or æ yo ö ÷ ç α′ = 2 ç ÷ æ y o ö è y o − y′ ø ç ÷ ç In 11. the result is the expression β′ = æ yo ö ç ÷ ÷ 2 ç ′ y y − o è ø æ ö yo ç ÷ In 11 .109) where W is the top width of the section. The relations are shown in Figure 2.107 and integrate.11 k ÷ s ø è 1 3 ì ïæ y o ö íç ç In y ′ ÷ ÷ −3 ï è ø î 2 æ yo ö æ yo ö 6y ′ ü ï ç ÷ ç ÷ In 6 In 6 − + + ý ç y′ ÷ ç y′ ÷ yo ï è ø è ø þ (2.106 and 2.13) Equation 2.108) These equations (Equations 2.106) Similarly. w.

12.30 .13. Figure 2. The river cross section.Figure 2. 2. Energy and momentum coefficients for a unit width of river.

Now.12 we are in a position to compute α and β for any river cross-section given the discharge measurement notes.109 can be written α= A2 W α ′ V 3 ( y o − y ′) dw 3 ò 0 Q (2. It is important to recognize that for river flows over floodplains. the term y′ is very small compared to yo so α= A2 W 3 ò 0 α ′ V y o dw Q3 (2.14 (SI) and 2. the expression for β is β= A ′ i å β i Vi ∆Q i Q2 (2.114) Now.110) Here α′ is the energy coefficient for the vertical section dw wide and yo deep.The second integral term in brackets in Equation 2.31 .111) Except for cases of low flow in gravel bed rivers. Equation 2.112) The discharge at a river cross-section is determined in the field by measuring the local depth and two local velocities at each of approximately 20 vertical sections.15 (English). 2. A calculation example is presented in Section 2. the correction factors α and β can be significantly larger than α′ and β′.114. In accordance with this general stream gaging procedure.2.112 could be written as α= or α= A2 2 Σ α′ ∆Q i i vi 3 i Q (2. V is the depth-averaged velocity in this vertical section and y′ = ks/30.113 and 2. with Equations 2.109 can be written y 3 3 ò y′ v dy = α ′ V ( y o − y ′) o (2.113) A2 ′ 3 i å α i v i y oi ∆w i Q3 Here. and Figure 2. the subscript i refers to the i-th vertical section. Equation 2. and ∆Qi is the river discharge associated with the i-th vertical or ∆Q i = Vi y oi ∆w i In a similar manner.

5.1 Gravity Waves The general equation for the celerity c (velocity of the wave relative to the velocity of flow) of a small amplitude gravity wave (ao << λ) is 2π y o ü ì gλ c = í tan h ý λ þ î 2π 1/ 2 (2.14.116) 2. only the basic one-dimensional analysis of waves and surges is presented. and (6) flood waves resulting from the progressive movement downstream of stream runoff or gradual release from reservoirs. Other aspects of waves are presented in other sections.14. Figure 2.0. (4) surges or bores resulting from sudden increase or decrease in the flow by opening or closing of gates or the movement of tides on coastal streams. (2) waves resulting from the surface instability that exists for flows with Froude numbers close to 1. Waves are an important consideration in bridge hydraulics when designing slope protection of embankments and dikes.32 . and channel improvements. 2.5 UNSTEADY FLOW Unsteady flows of interest to the designer of waterway crossings and encroachments are: (1) waves resulting from disturbances of the water surface by wind and boats. (5) standing waves and antidunes that occur in alluvial channel flow.0. For deep water waves (short waves) defined as yo 1 > λ 2 (2. (3) waves resulting from flow disturbance due to change in direction of flow with Froude numbers greater than about 2.115) where the terms are defined in Figure 2. Definition sketch for small amplitude waves. In the following paragraphs.2.

122) Generally as 2ao/yo approaches unity the crest develops a sharp peak and breaks. If the wave is moving opposite to the flow then. When V = c for small amplitude flow.115) reduces to ì gλ ü c=í ý î 2π þ 1/ 2 (2.119. when c > V.123) The ratio of the flow velocity to the celerity of a shallow water wave of small amplitude is defined by the Froude number: 2. In Equation 2.117.118) The time of travel of one water crest to another at a given point is called the period T and can be defined from the celerity and wave length c = λ/T (2. In the above equations. and is given by the expression ì ( y + 2a o ) 2 ü c=í o gy o ý î(y o + ao ) y o þ 1/ 2 (2. the celerity is independent of depth and depends on gravity g and wave length λ. This is the celerity of ocean waves. the waves move upstream. The celerity of finite amplitude shallow water waves has been determined both analytically using Bernoulli's equation and experimentally.120) In Equation 2.121) When 2ao is small in comparison to yo ì 2a ü c = í1 + o gy o ý yo î þ 1/ 2 (2.the celerity relationship (Equation 2. V = c = gy o (2. the wave is stationary.33 . when c = V. the celerity is a function of gravity and depth which describes small amplitude waves in open channels.117) For shallow water waves (long waves) yo 1 < λ 20 Then Equation 2. that is ao/λ << 1. the wave moves downstream. c is measured relative to the fluid.119) (2. and when c < V.115 reduces to c = gy o (2. These two equations apply only to small amplitude waves.

124) When Fr < 1 (subcritical or tranquil).34 . or from an incoming tide.5. 2. A surge may result from sudden release of water from a dam. the first wave breaks and produces a discontinuous surface. or of a surge that moves downstream as the result of a sudden opening of a gate. As it moves upstream a negative surge quickly flattens out. The fact that waves or surges cannot move upstream when the Froude number is equal to or greater than 1.15. Equation 2. However. a small amplitude wave moves upstream. Equation 2. it also causes a negative surge to move upstream. The breaking wave dissipates energy and the previous equations for wave celerity are invalid. 2. The lifting of a gate in a channel sketched in Figure 2. by applying the momentum and continuity equations for a control volume encompassing the surge shown in Figure 2. the equation for the celerity of a surge can be derived for flat slopes as: ì é1 y2 æ y2 öù ü ï ï ç c = így 1 ê + 1÷ ÷ú ý ç 2 y y ï ê 1 è 1 øú ë ûï þ î 1/ 2 (2. the surge has an undulating wave form.125) Figure 2. a small amplitude wave moves downstream and when Fr = 1 (critical flow).125 gives the velocity of a surge as it moves upstream as the result of a sudden total or partial closure of gates. Sketch of positive and negative surges. When Fr > 1 (supercritical or rapid flow). If 2ao/yo is greater than one.125 is approximately correct for the celerity of the negative surge if the height of the surge is small compared to the depth. The Froude number is not only the ratio of the flow velocity to the celerity of a shallow water wave. but is also the ratio of the inertia forces to the gravity forces. or of an incoming tide.15.2 Surges A surge is a rapid increase in the depth of flow.0 is important to remember when determining when the stage-discharge relation at a cross-section can be affected by downstream conditions. If the ratio of wave height 2ao to the depth yo is less than unity. a small amplitude wave is stationary.15b not only causes a positive surge to move downstream.Fr = V gy o (2.

Bureau of Reclamation (Chow 1959) classifies the hydraulic jump on a flat slope into various types as illustrated in Figure 2. They travel at velocities greater than the normal flow and grow in size as they progress downstream. These waves and jets can cause erosion a considerable distance downstream of the jump.5. or force the operation of the channel at inefficient discharges in order to prevent damage. The results of these experiments are given in Figure 2. 2. the energy dissipation of the jump is not large and jets of high velocity flow can exist for some distance downstream of the jump. For additional information on hydraulic jumps. 2. When the Froude number for rapid flow is less than 1. for example. Chow (1959) and Rouse (1950).125 is the equation for a hydraulic jump on a flat slope. The U.17. they may cause considerable damage.18).126) or y2 1 1/ 2 ( = 1 + 8Fr12 ) − 1 y1 2 { } (2.16 is recommended. when the Froude number of the approaching flow is less than three. In addition.2. It can be shown that this head loss is: hL = ( y 2 − y 1 )3 4y1 y 2 (2. Roll waves can be superposed over the normal flow in an open channel. Such flow is not at all uncommon with harmless thin sheets of flow on sloping sidewalks.16. Solutions for a hydraulic jump on a sloping channel are given in HEC-14. the rate of energy dissipation in the jump is very large and Figure 2.128 has been experimentally verified along with the dependence of the jump length Lj and energy dissipation (head loss hL) on the Froude number of the approaching flow.35 .7. surges of an intermittent nature may occur which are called roll waves or slug flow (Figure 2.4 Roll Waves Under certain conditions on steep slopes.S. When these roll waves occur in large open channels. When V1 equals the celerity c of the surge the jump is stationary and Equation 2.127) The corresponding energy loss in a hydraulic jump is the difference between the two specific energies. Equation 2. however.125 can be rearranged to the form: V1 ì öü ï1 y2 æ y2 ï ç = Fr1 = í + 1÷ ý ÷ ç 2 y y ï gy 1 1 è 1 øï î þ 1/ 2 (2. see HEC-14.128) Equation 2. The waves are propagated for a considerable distance downstream. For larger values of the Froude number. an undulating jump with large surface waves is produced.3 Hydraulic Jump When the flow velocity V1 is rapid or supercritical the surge dissipates energy through a moving hydraulic jump.5.

2.Figure 2. Various types of hydraulic jump (Chow 1959). Hydraulic jump characteristics as a function of the upstream Froude number.17.16.36 . Figure 2.

When the flow in a wide channel is turbulent with a smooth boundary.133) These conditions indicate that.5 (2. roll waves can form in laminar flow in a wide channel if the velocity is half the celerity of a gravity wave. the Froude number is greater than 2.18. Roll waves or slug flow. S is the slope of the channel.5 times the celerity of a wave. a necessary condition required to generate instability of the free surface and induce the formation of roll waves in turbulent flows when Chezy's equation is applicable is: Fr = V gy o >2 (2. since their size depends upon the magnitude of the discharge. roll waves can occur when the flow velocity is greater than twice the celerity of a wave that is. or the slope is 2. the flow may never pass through critical flow (Fr = 1. By way of contrast. However. the length of the channel.25 times the slope required for critical depth.129) which can be expressed in alternate form for a wide channel as S≥4 g C2 (2. They can also form for turbulent flow in a wide channel with a smooth boundary if the velocity of flow is greater than 1. the type of flow (laminar or turbulent). 2.Figure 2. for turbulent flow in a wide channel with a rough boundary. roll waves can form if Fr ≥ 0. or when the slope is four times as great as the slope required for critical depth.5 S ≥ 2.132) and when the flow in a wide channel is laminar. and C is the Chezy discharge coefficient. There is no simple criterion for determining the size of roll waves. the roughness and slope of the channel.130) for turbulent flow with a rough boundary in which yo is the normal depth.0). and the nature and frequency of the initial disturbances which cause the waves to form. roll waves can form if Fr ≥ 1.37 .25 g C2 (2. in other words.131) (2.

Refinement of the analysis can be made in a second step by including friction loss (see HEC-14.134) 2.38 . Transitions in open channel flow. to increase or decrease bottom elevation (Figure 2.19. The first step in the analysis is to use the Bernoulli equation (neglecting any head loss resulting from friction or separation) to determine the depth and velocity changes of the flow through the transition.6 STEADY RAPIDLY VARYING FLOW 2. blockage. Transitions are used to contract or expand a channel width (Figure 2.2. the water surface elevation through a transition is determined using the Bernoulli equation and then modified by determining the friction loss effects on velocity and depth in short reaches through the transition. any time a transition changes velocity and depth such that the Froude number approaches unity. The Bernoulli equation for flow in Figure 2. Contracting flows (converging streamlines) are less susceptible to separation than for expanding flows. Further refinement depends on importance of freeboard.19b). If the approaching flow is supercritical.6. or choking of the flow may occur. at least as a first approximation.19a). whether flow is supercritical or approaching critical conditions. Figure 2. Energy losses resulting from flow separation cannot be neglected. or to change both the width and bottom elevation.19b is: V12 V2 + y 1 = 2 + y 2 + ∆z 2g 2g or (2.1 Flow Through Transitions Steady flow through relatively short transitions where the flow is uniform before and after the transition can be analyzed using the Bernoulli equation. Unsubmerged flow through bridges or culverts can be considered as flow through transitions. For example. Also. Chapter 4). a hydraulic jump may result. Energy loss due to friction may be neglected. and transitions where separation may occur need special treatment which may include model studies. problems such as waves.

the following specific energy (often referred to as specific head) analysis is done on a unit width of channel so that Equation 2.137) For a given q. dH q2 = 1− 3 = 0 dy gy (2.136) The term H is called the specific head. Flow cannot occur with specific energy less than the minimum. Thus.136 becomes: H= q2 +y 2gy 2 (2. To determine yc the derivative of H with respect to y is set equal to 0. and is the height of the total head above the channel bed. There are two possible depths called alternate depths for any H larger than a specific minimum.6. Figure 2.137 can be solved for various values of H and y. Figure 2. When y is plotted as a function of H. The single depth of flow at the minimum specific energy is called the critical depth yc and the corresponding velocity.138) 2. the critical velocity Vc = q/yc. for specific energy larger than the minimum. Specific energy diagram.135) (2.20.2 Specific Energy Diagram For simplicity.39 .20 is obtained (Rouse 1946).H1 = H 2 + ∆z where H= V2 +y 2g (2. Equation 2. the flow may have a large depth with small velocity or small depth with large velocity. 2.

140) Note that Vc2 = y c g or Vc gy c but V gy also Hmin Vc2 3 = + y c = y c (see Equation 2. Figure 2.147) 2g 2 (2.139) =2 Vc2 (see Equation 2.141) Thus.143) =1 (2. Equation 2. By plotting y as a function of q. for a given flow in a channel.22 is obtained and for any discharge smaller than a specific maximum.40 .3 Specific Discharge Diagram For a constant H. two depths of flow are possible (Rouse 1946).147) 2g (2. Thus. Flows with velocities larger than critical (Fr > 1) are called rapid or supercritical and flow with velocities smaller than critical (Fr < 1) are called tranquil or subcritical. flow at minimum specific energy has a Froude number equal to one. These flow conditions are illustrated in Figure 2. 2.142) (2.6.144) = Fr (2. 2.and q = (gy 3 c) or æ q2 ö yc = ç ç g÷ ÷ è ø 1/ 3 1/ 2 (2.137 can be solved for y as a function of q. a rise in the bed level can occur up to a ∆zmax without causing backwater. where a rise in the bed causes a decrease in depth when the flow is tranquil and an increase in depth when the flow is rapid. If the rise in the bed is increased beyond ∆zmax for Hmin then the approaching flow depth y1 would have to increase (increasing H) or the flow would have to be decreased. Furthermore.21. there is a maximum rise in the bed for a given H1 where the given rate of flow is physically possible.

21.145) 2.Figure 2.137 is rearranged to obtain: q = y 2g(h − y ) The differential of q with respect to y is set equal to zero. Changes in water surface resulting from an increase in bed elevation. q ( 2H − 3 y ) dq =0= dy 2 (H − y )1 / 2 (2.41 . Figure 2.146) (2. Specific discharge diagram.22. To determine the value of y for qmax Equation 2.

That is. and expansion. Flow conditions for constant specific head for a width contraction are illustrated in Figure 2.23 assuming a fixed bed and no geometrical effects such as eccentricity.0.42 . the Froude number is 1. skew. 2. A further decrease in width will cause backwater. The maximum possible contraction without causing backwater effects occurs when the Froude number is one. the discharge per foot of width q is a maximum. the discharge is a maximum for a given specific head and the specific head is a minimum for a given discharge.149) For critical conditions. Figure 2. piers. The contraction causes a decrease in flow depth when the flow is tranquil and an increase when the flow is rapid.0.148) 2Vc2 2 H= 3 2g (2.23.from which yc = or Vc = gy c (2. From this: 2 2Vc2 æ q max 2 yc = H = =ç ç g 3 2g è ö ÷ ÷ ø 1/ 3 (see Equation 2. and the flow is critical.140) (2. the Froude number is 1. and yc is 2H/3. an increase in depth upstream will occur to produce a larger specific energy and increase yc in order to get the flow through the decreased width.147) Thus for maximum discharge at constant H. Change in water surface elevation resulting from a change in width.

Rouse et al.6. For an expansion. for practical purposes. until the flow could go through the constriction and establish uniform flow downstream.4 Transitions With Super Critical Flows Contractions and expansions in rapid flows produce cross wave patterns similar to those observed in curved channels (Ippen 1950 and Chow 1959). the divergent walls are followed by a transition to parallel lines. then the flow could not be sustained with this amount of energy loss. Therefore. If a hydraulic jump is acceptable. A satisfactory straight transition can be created by flaring the walls so that tan θ = 1/3 Fr. Figure 2. 2. the upstream flow would follow a natural streamline to a vena contracta. This criteria recommended by Blaisdel (1949) avoids creating an abrupt expansion. The cross waves are symmetrical with respect to the centerline of the channel.150) where W is the channel width and the subscripts 1 and 2 refer to sections upstream and downstream from the contraction. 2. A model study should be used to determine transition geometry where a hydraulic jump is not desired. Ippen and Dawson (1951) have shown that in order to minimize the disturbance downstream of a contraction.43 . (1951) found experimentally that the most satisfactory boundary form is given by: w 1æ x ö ÷ = ç ÷ W1 2 ç è W1 Fr1 ø 3/2 + 1 2 (2. This requires a long transition and should not be attempted unless the structure is of primary importance. the inlet structure can be designed using the procedure in HEC-14. There would be interaction of the flow in the separation zone and considerable energy would be lost. Backwater would occur. An increase in slope of the bed downstream from C and no separation would allow the flow to follow the line A to C to E. the length of the contraction should be: L= W1 − W2 2 tan θ (2. but then downstream.23 is drawn with the side boundary forming a smooth streamline. If the contraction were due to bridge abutments. If the slope downstream of the abutments was the same as upstream. Similarly the flow can go from B to C and back to E or up to D depending on boundary conditions. A boundary developed from this equation diverges indefinitely. The contraction angle is θ and should not exceed 12°. increasing the depth in the constriction and upstream.The flow in Figure 2. Chapter 4B. Tranquil approach flow could follow line A-C but the downstream flow probably would not follow either line C-D or C-E but would have an undulating hydraulic jump. the flow would probably separate.151) where x is the longitudinal distance measured from the start of the expansion or outlet section and w is the lateral coordinate measured from the channel centerline.23 can go from point A to C and then either back to D or down to E depending on the downstream boundary conditions.

If the flow impinges on an erodible bed it will scour a deep hole.154) y1 h = 0.24.6. Flow geometry and conditions of interest are: • • • • • • discharge for unit width.5 Flow Over Drop Structures Subcritical flow over vertical wall drop structures at flat slopes is illustrated in Figure 2.27 h 2 (2. q drop height.00 D 0. Ld depth and velocity at impingement.44 .425 = 1.156) y2 h 2.66 D 0.153) yp h = 1.22 (2. Depending on downstream conditions (bed elevation and slope and down stream flow depth) the flow will continue to be supercritical or go through a hydraulic jump. Flow characteristics over a drop structure (Chow 1959).2.24. Chow (1959) gives equations to determine the geometric terms as functions of a drop number D. y1. yp sequent depth if a jump forms.30 D 0.152) (2. As illustrated. the flow impinging on an apron will continue downstream but at supercritical depth and velocity. h distance from drop wall to point of impingement.54 D 0. V1 pool depth under nape.155) (2.27 (2. Methods to calculate the depth of this scour hole are given in Chapter 3. In English units of measurement these are: q D= g h3 LD = 4. y2 Figure 2.

proceeds at an approximately equal rate along the length of the river.. The equations are for a well-aerated nappe.7. For upstream subcritical flow the brink serves as the control and the subcritical approach flow goes through critical 3 to 4 yc upstream of the brink (Chow. 2.e.25. Figure 2. 2. If the tailwater depth. incomplete from the standpoint of the work of the river and of the physical nature of this phenomenon. the jump will move upstream and the jump will be submerged. It is. The first type includes those in which the river bends follow the curves of the valley so that each river bend includes a promontory of the parent plateau. y2.1 Types of Bends Two principal types of bends are deepened or entrenched bends and meandering surface bends. If the tailwater depth. Such meandering when not disturbed by the influence of external factors.45 . For upstream supercritical flow the control for the flow is upstream and the profile for the nappe is a function of the approach Froude number. If the nape is poorly or not aerated. This division of bends is correct and sufficiently definitive with respect to external forms of the relief and the processes of formation and development of bends. 5. alluvium covered valley floor. Ippen (1950. Both of these types of bends can be put into one category--the category of freely meandering channels. 3.7 FLOW IN BENDS 2. the hydraulic jump will move downstream.Note: 1. Nappe profiles for supercritical flow (Ippen 1950). 2. p. p. 4.25). 44). 1959. The second type includes bends which are formed only by the river on a flat. i. and where the slopes of the valley are not involved in the formation of such bends. 533) gives a composite plot of nappe profiles for different values of the Froude number (Figure 2. the nappe will move upstream until in contact with the vertical wall. is less than yc. is larger than yc. meandering determined only by the interaction of the stream and the bed material. however. y2.

In the immediate vicinity of the banks. The radius of curvature increases with the density of the material. a cross section through a typical bend is shown. (2) limited bends 7. The limited bends have the greatest radii. Forced bends .0. In Figure 2. As shown in Figure 2. and (3) forced bends 2. one can distinguish the following three types of bends of a natural river channel: 1. where there appears to be a concentrated deep scour hole. A typical feature of bends is a close relationship between the type of stream bend and the radius of curvature. In all cases. the depth gradually increases and the maximum depth is found some distance below the apex of the bend.0 to 8.Both banks are composed of alluvial floodplain material which is usually quite mobile.26a. In forced bends the greatest depth is located in the middle third of the bend.7. Limited bends are entrenched bends. Considering both the action of the stream and the interaction between the stream and the channel. a third type of bend is often encountered.5 to 3. the centrifugal force is greater near the surface where the fluid velocity v is greater and less at the bed where v is small. the sum of the centrifugal and excess hydrostatic forces varies with depth and can cause a lateral velocity component. Limited bends . 2. A second characteristic feature of bends is the distribution of depths along the length of the bend. and 3.0. The magnitude of the transverse velocity is dependent on the radius of curvature and on the proximity of the banks. 2. The differential hydrostatic force is uniform throughout the depth of the control volume. 2.The banks of the stream are composed of consolidated parent material which limits the lateral erosion by the stream. Free bends . to a certain degree. and the differential hydrostatic force γdz caused by the superelevation of the water surface dz. This bend occurs when the stream impinging on a erosion-resistant bank forms a forced curve which is gradually transformed into a river bend of a more constricted shape.Under natural conditions. Next in size are the radii of free bends. The average values of the ratios of the radii of curvature to the width of the stream at bankfull stage for the three types of bends are: (1) free bends 4. In free bends and limited bends.The stream impinges onto an almost straight parent bank at a large angle (60° to 90°). the free bend corresponds to the common concept of a surface bend. determines the radius of curvature of the channel in a free bend.0. In the forced bend.46 . as well as the general laws of their formation.2 Transverse Velocity Distribution in Bends The transverse velocities in bends result from an imbalance of radial pressures on a particle of fluid traveling around the bend.26b.5 to 5. The radial forces acting on the shaded control volume are the centrifugal force mv2/r in which r is the radius of curvature. the depth sharply increases at the beginning of the bend and then gradually diminishes. The forced bend has the smallest radius of curvature. there can be no lateral velocity if the river is narrow and deep. As shown in Figure 2.26. the effect of the character (density) of the bank material is important and.

the differential pressure in the radial direction arises from the radial acceleration or: 2 1 ∂p Vθ = ρ ∂r r (2.3 Subcritical Flow in Bends Because of the change in flow direction which results in centrifugal forces.. Definition sketch of flow around a bend.e. 2. 2. Using cylindrical coordinates (Figure 2.27. there is a superelevation of the water surface in river bends. The resulting transverse slope can be evaluated quantitatively.157) To calculate the total superelevation between the outer and inner bend two assumptions are made: (1) radial and vertical velocities are small compared to the tangential velocities such that Vθ ≈ V.7.26.47 .Figure 2.27). p = γy. and (2) pressure distribution in the bend is hydrostatic. i. The water surface is higher at the outside bank than at the inside bank. Schematic representation of transverse currents in a channel bed. Figure 2.

48 . With this assumption and assuming a constant average specific head.161) in which C1 = rV is the free vortex constant. Woodward (1920) assumed V equal to the average velocity Q/A and r equal to the radius to the center of the stream rc. Ippen and Drinker (1962) reduced Equation 2. By assuming flow depth of flow upstream of the bend equal to the average depth in the bend. and obtained: ∆Z = or ∆Z = z o − z i = V2 (ro − ri ) g rc (2.Then ∆Z = 1 ro V 2 dr òr g i r (2.158) To solve Equation 2.160) 1 ro V 2 dr òr g i rc (2.162) For situations where high velocities occur near the outer bank of the channel. The results obtained assuming various velocity distributions follow. and zo and ro are the water surface elevation and the radius at the outside of the bend.158. Ippen and Drinker (1962) obtained: 2. a forced vortex may approximate the flow pattern.161 to: ì ï V 2 2W ï 1 ∆Z = í 2g rc ï æ W ï1 − ç ç î è 2rc ü ï ï 2 ý ö ï ÷ ÷ ï ø þ (2. Shukry (1950) obtained: 2 2 C1 1 ro C1 ∆Z = ò ri 3 dr = g 2g r ì 1ü ï1 ï í 2 − 2ý ï ro ï î ri þ (2. the transverse velocity distribution along the radius of the bend must be known or assumed. By assuming the velocity distribution to approximate that of a free vortex (Vθ = C1/r).159) in which zi and ri are the water surface elevation and the radius at the inside of the bend.

Therefore. One method is to bank the floor of the channel and the other is to provide curved vanes in the flow. superelevation should be computed using Equations 2. it is recommended that Equation 2. The waves produced form a series of troughs and crests in the water surface along the channel walls. the turning effect of the walls is not felt on all filaments of the flow at the same time and the equilibrium of the flow is destroyed. These disturbance patterns are the result of non-equilibrium conditions which persist because the disturbances cannot propagate upstream or even propagate directly across the stream. and in alluvial channels the resulting erosion of the outside bank and deposition on the inside bank leads to further error in computing superelevation.7. respectively. 2. C 2 = Vmax and C1 = Vmax rc rc and Equation 2.164) and when r = rc.162 or 2.158 be used to compute superelevation in alluvial channels. For lined canals with strong curvature. An example showing how to calculate superelevation in bends from velocity measurements is presented in Section 2. then: ∆Z = 2 2 1 ro C1 1 rc C 2 2 r + dr dr òr r òr g c r3 g i (2. Two methods have been used in the design of curves for rapid flow in channels.4 Supercritical Flow in Bends Rapid flow or supercritical flow in a curved prismatic channel produces cross wave disturbance patterns which persist for long distances in a downstream direction.14 (SI) and 2.163) By assuming that the maximum velocities are close to the centerline of the channel in the bend and that the flow pattern inward and outward from the centerline can be represented as forced and free vortices. Banking on the floor produces lateral forces which act simultaneously on all filaments and causes the flow to 2.é ê V 2 2W ê 1 ∆Z = ê 2g rc æ W2 ê1 − ç 2 ç ê ë è 12rc ù ú ú öú ÷ú ÷ú øû (2. The example also compares the various approximate equations included in this section.49 .165. Therefore.15 (English) at the end of this chapter.164 becomes: ∆Z = 2 Vmax 2g ì æ ri ï í2 − ç çr è c ï î 2 2 ö æ rc ö ü ï ÷ ç ÷ − ÷ çr ÷ ý ø è oø ï þ (2. V = Vmax Therefore.165) The differences in superelevation that are obtained by using the different equations are small.

resistance is neglected and changes in depth due to acceleration are dominant.0. changes in cross-section. each small channel has a smaller superelevation. Reasons for the depth being different than the normal depth are changes in slope of the bed. which are often called backwater curves. the slope of the water surface and the slope of the energy gradeline are all parallel and are equal to the head loss divided by the length of the channel in which the loss occurred. In uniform flow. There are no changes in cross-section or flow direction. and the depth (called normal depth) is constant. the actual flow depth y is either larger or smaller than the normal depth yo and either larger or smaller than the critical depth yc. changes in cross-section. the first step is to determine of the general characteristics of the water surface and what type of backwater curve would exist. This type of flow is called gradually varied flow. Curved vanes break up the flow into a series of small channels and since the superelevation is directly proportional to the channel width. In most cases a physical model study is recommended. In this type of flow. The total head is: 2. The water surface profiles. The second step is to perform the numerical computations to determine the elevation of the water surface or depth of flow. There are design methods in the literature (Rouse 1950. In working with gradually varied flow. 2. depend on the magnitude of the actual depth of flow y in relation to the normal depth yo and the critical depth yc. In rapidly varied flow through short streamlined transitions. acceleration forces are not zero. Chow 1959).turn without destroying the flow equilibrium.1 Introduction Thus far. Different conditions prevail for each of these two types of steady flow. acceleration forces are zero and energy is converted to heat as a result of viscous forces within the flow.2 Classification of Flow Profiles The classification of flow profiles is obtained by analyzing the change of the various terms in the total head equation in the x-direction. resistance to flow dominates and acceleration forces are neglected. obstruction to flow and imbalances between gravitational forces accelerating the flow and shear forces retarding the flow. two types of steady flow have been considered. and the critical depth is the depth of flow when the Froude number equals 1. and viscous forces can be neglected (at least as a first approximation). Normal depth yo is the depth of flow that would exist for steady-uniform flow as determined using the Manning or Chezy velocity equations. or depth take place in relatively short distances.8. the slope of the bed. If the bend is not properly shaped or designed a hydraulic jump may occur or the cross-waves can be amplified. Ippen 1950. a third type of steady flow is considered. 2. changes in depth and velocity take place slowly over large distances. They are uniform flow and rapidly varied nonuniform flow.8 GRADUALLY VARIED FLOW 2. In gradually varied flow.8.50 . In this section. In rapidly varied flow. direction. In steady uniform flow.

167) Then assuming a wide channel for simplicity dH T q 2 dy dy dz + 3 = + dx gy dx dx dx (2.169) The term dy/dx is the slope of the water surface Sw.84) So = − q2 2 Co y3 o (2.169 and 2. and dz/dx is the bed slope -So.168) The term dHT/dx is the slope of the energy gradeline Sf. For short distances and small changes in y the energy gradient can be evaluated using the Manning or Chezy velocity equations. Equation 2.166) (2.51 .170) where the subscript "o" indicates the steady uniform flow values. When Equations 2.171) If Manning's equation is used to evaluate Sf and So.170 are substituted into Equation 2. the familiar form of the gradually varied flow equation is obtained: ì æ C ö2 æ y o ï1 − ç o ÷ ç ç dy ï è C ø è y = So í 3 dx æ yc ö ï ç ÷ − 1 ç y ÷ ï è ø î 3 ö ü ÷ ÷ ï ø ï ý ï ï þ (2.172) 2.HT = or HT = V2 +y+z 2g Q2 +y+z 2gA 2 (2. For steady flow. by assumption.168. the bed slope is (from Equation 2. When Chezy's equation (Equation 2.171 becomes ì æ ö2 yo ï1 − ç n ÷ æ ç ç ç ÷ ï no è y dy ï = So í è ø 3 dx æ yc ö ï ï 1− ç ÷ ç y ÷ è ø ï î ö ÷ ÷ ø 10 / 3 ü ï ï ï ý ï ï ï þ (2.84) is used the expression for dHT/dx is − dH T q2 = Sf = 2 3 dx C y (2.

29. the ratio of the normal depth yo to the actual depth y and the ratio of the critical depth yc to the actual depth y. Figure 2. and then sketching the backwater curves.28 and summarized in Table 2. In general. the qualitative analysis of the flow profile depends on locating the control points. the flow cannot become tranquil without a hydraulic jump occurring. Equations 2. tranquil flow can become rapid (cross the critical depth line). With n = no. When y → yc.52 . when the flow is rapid (Fr ≥ 1). These are illustrated in Figure 2. determining the type of curve upstream and downstream of the control points. The backwater curves that result from a change in slope of the bed are illustrated in Figure 2.172 indicate that dy/dx is perpendicular to the bed slope when y → yc. the depth control is downstream and the computations must proceed upstream. there are twelve types of water surface profiles. When analyzing long distances 30 to 300 m (100 to 1. When there is a change in cross-section or slope or an obstruction to the flow. the control of the depth is upstream and the backwater proceeds in the downstream direction. It must be remembered that when flow is rapid (Fr > 1). The difference between flow resistance for steady uniform flow no to flow resistance for steady nonuniform flow n is small and the ratio is taken as 1.3.000 ft) or longer one can assume qualitatively that y reaches yc.171 and 2.The slope of the water surface dy/dx depends on the slope of the bed So. 2. the assumption that acceleration forces can be neglected no longer holds.29. This is illustrated in Figure 2. Classification of water surface profiles.28.0. In contrast. For cross-sections close to the cross-section where the flow is critical (a distance from [3 to 15 m (10 to 50 ft)]). curvilinear flow analysis and experimentation must be used to determine the actual values of y. When flow is tranquil (Fr < 1).

So equals Sc and yo = yc. the actual depth of flow y is greater than both the normal depth yo and the critical depth yc. For a critical slope. control of the flow is downstream. So is smaller than Sc and yo > yc. For a mild slope.28. S2. So is larger than Sc and yo < yc. The flow is tranquil for M2. See paragraph preceding Figure 2. Class Mild Mild Mild Critical Critical Steep Steep Steep Horizontal Horizontal Adverse Adverse Note: (1) With a type 1 curve (M1. Flow is rapid for S2 and the control is upstream. For a steep slope. S3. The case where y → yc is of special interest. With a type 3 curve (M3.3. For a horizontal slope. So is negative. C1). A3. A2. the actual depth y is smaller than both the normal depth yo and the critical depth yc. the actual depth y is between the normal depth yo and the critical depth yc.Table 2. For an adverse slope. Because flow is tranquil. H2). So equals zero. H3). S1. Bed Slope So>0 So>0 So>0 So>0 So>0 So>0 So>0 So>0 So = 0 So = 0 So<0 So<0 Depth y>yo>yc yo>y>yc yo>Yc>y y>yo=yc y<yo=yc y>yc>yo yc>y>yo yc>yo>y y>yc yc>y y>yc yc>y Type 1 2 3 1 3 1 2 3 2 3 2 3 Classification M1 M2 M3 C1 C2 S1 S2 S3 H2 H3 A2 A3 (2) (3) (4) (5) (6) (7) (8) (9) 2. With a type 2 curve (M2. The flow is rapid and control is upstream. C3. Characteristics of Water Surface Profiles. A2 and H2 and thus the control is downstream.53 .

2.Figure 2. Examples of water surface profiles.29.54 .

175) (2. 2.176. The standard step method is derived from the energy equation V12 V2 + y 1 + ∆z = 2 + y 2 + HL 2g 2g From Figure 2.8. Definition sketch for the standard step method for computation of backwater curves.55 . It is recommended that the assumed depth be kept relatively close to the known value of y1 in order to keep the interval ∆L as short as possible to obtain better accuracy in the calculation.30.2.174) (2. assume another y either upstream or downstream depending on whether the flow is tranquil or rapid.8. Prior knowledge of the type of backwater curve as classified in Section 2.3 Standard Step Method for the Computation of Water Surface Profiles The standard step method is a simple computational procedure to determine the water surface profile in gradually varied flows.176) (2. which can be obtained from a stage-discharge relationship. The procedure is to start from some known y. and compute the distance ∆L to the assumed depth using Equation 2.30 V2 V12 + y 1 + S o ∆L = 2 + y 2 + S f ∆L 2g 2g H1 + S o ∆L = H 2 + S f ∆L and ∆L = H 2 − H1 So − S f (2.2 is useful to determine whether the analysis should proceed upstream or downstream.173) Figure 2.

study of the rainfall records and discharge records of other gages in the drainage basin. but the stage discharge relation was poorly defined with no consistent curve. the magnitude of the sediment transported by the stream. These terms are defined in the glossary and will be defined later in this section. Thus.2 Gaging Station A gaging station consists of structures and equipment to measure and record the stage and discharge as a function of time at a given site on a stream.300 cfs).800 to 40. 2. analyses of the longterm stream flow data are made (flood frequency. only channel control. The stage record for the 1987 storm was excellent. The most extensive records of streamflow are by the U. and (3) computing. For example. The streambed was sand and there was no bed rock control. Stage or gage height is the height of the water surface above a chosen arbitrary datum corresponding to the zero of the gage. trends). Often a gaging station will be located near the site of a bridge crossing. The gage height record was good.300 cfs) was determined from the slopearea measurement.9 STREAM GAGING 2.140 m3/s (14. the stage-discharge relation was well defined with a single curve established with actual discharge measurements made over many years and large flows. The range in discharge for the slope area measurement resulted from assumptions on Manning's n and the amount of degradation. A program to obtain a systematic record of the stream flow consist of (1) establishing and constructing a streamflow measurement station. State.56 . The gage height record was excellent with few time gaps. These depend on stream characteristics (bed material. In the following sections a typical gaging station.9. In contrast. in many cases. regional or other hydrologic tools can be used to analyze the stations in the region to obtain the hydrology for the bridge site.b.5 km (14 mi) upstream of the bridge.9. Geological Survey. In addition. The peak discharge for the flood had to be determined using indirect methods with results that ranged from to 420 to 1. The zero of the gage is related to sea-level elevation. discharge measurement. low flow analysis. If not. Buchanan and Somers 1968a. The discharge used in the final analysis 773 m3/s (27. Their records of stream flow and analysis can be obtained for any State at their District offices or from their Web Site. The methods of obtaining water discharge records are described in publications of the USGS (Corbet 1962.000 sites in the United States in cooperation with other Federal. the hydrology used for the analysis of the 1987 I-90 bridge failure in upstate New York was excellent. Carter and Davidian 1968. The USGS has been collecting stream gaging records since 1888 (Corbet 1962) and maintain over 4. The quality of the hydrologic records at a stream gaging site depends on obtaining an accurate gage height record and stage-discharge relation. An accurate record of stage is essential for computing the discharge for any period of time. Discharge is the rate of flow in m3/s or cfs. The discharge was routed to the bridge from the gage site using the U. and Cities. Counties. and Kennedy 1983).S.1 Introduction The hydrology of the stream at a bridge crossing is determined from records obtained by State and Federal agencies.S. the hydrology for the 1997 I-5 bridge failure in California was not well defined. There was a gaging station located 22. The stream at the gaging site was in bedrock with an excellent control. cross-section. stage discharge relation and the determination of the daily discharge will briefly be explained. compiling and publishing the stream flow data. and control) and.2. (2) operating and maintaining the station. 2. Corps of Engineers HEC-1 model to obtain the hydrology of the flow at the bridge.

however. Discharge is measured using a current (velocity) meter by wading at low flow and from a cableway or highway bridge at high flows. a manometer and gas under pressure bubbling into the stream are used to measure stage instead of a float. Gaging station well and shelters (from Buchanan and Somers 1968a). Figure 2. Figure 2. Also.34 illustrates a stage versus time chart. The stage versus time is recorded by pen on paper. or both. at many stations the stage is transmitted over phone lines or satellite to a central location.33). The procedure for an actual current meter measurement is described in the next section.Typical gaging stations are illustrated in Figure 2.31. well and pipe intakes (Figure 2. In many cases.57 . or electronically on tape.31 and 2.32. large flows are measured by indirect methods such as a slope-area measurement using Manning's equation. In many gaging stations. the preferred method is to use a current meter measurement. 2.

Typical float recording gaging station (from Buchanan and Somers 1968a).32. Typical bubble and manometer recording gaging station (from Buchanan and Somers 1968a).58 .33. 2. Figure 2.Figure 2.

59 . Stage vs.Figure 2.34. time hydrograph (Kennedy 1983). 2.

5 ft) the velocity is the average of the velocity measured at 0.5 ft) the velocity is measured at 0.9.3 Measuring Water Discharge Generally the stream cross-section is divided into 20 sections (verticals).8 m (2. If over 0. The sub-area is determined by multiplying the measured depth by 0.8 the depth below the water surface. and the average velocity determined. In complex velocity distributions.2 and 0.6 the depth below the water surface. 2. plotted.5 times the width to the adjacent sections (Figure 2.35.8 m (2.60 . two points.2. The average velocity in the vertical (one point. The velocity is measured using a vertical-axis rotor with cups (Price is typical). Total Discharge is the sum of all the unit discharges. At each vertical the depth and velocity are measured. and etc measurement) is multiplied by the sub-area to obtain a unit discharge. horizontal-axis rotor with vanes (Ott is typical) or a magnetic (Brien-Macnertry is typical) velocity meter.35). additional point velocities are taken. Figure 2. Definition sketch of computing area and discharge at a gaging station (from Buchanan and Somers 1968b). If the depth is less than 0.

4 Stage-Discharge Relation From the recorded stage and corresponding discharge measurement of many measurements over time at a gaging station. the daily discharge is determined using the average gage elevation for that day. Figure 2. and the latter from fair to poor. From the recorded stage-time record and the stage-discharge relation the daily. If there is a rapid change in stage with time.61 . a stage vs.36 is an example of an excellent well-defined stage-discharge relation for a station with a good control. discharge relation is developed. and minimum flow at the station are determined. yearly peak. Figure 2. daily discharge is determined by converting the timestage curve to time-discharge and by integrating over the 24 hours to determine the average daily discharge. An additional example of a sand channel stage-discharge relation is given in Chapter 3. The discharge for the first would be rated from good to excellent. Figure 2.36. 2. If for a given day the stage doesn’t change appreciably.2.9.37 is an example of a poorly defined stage-discharge relation for a gage with a sand channel and channel control. Stage-discharge relation for Schoharie Creek. New York (Butch 2000).

CA). Other concerns discussed in the following chapters include the stability and scour around embankments.Figure 2. 2.37. This loss of energy is reflected by a rise in the water surface and the energy line upstream from the bridge. and local scour around piers.62 . 2.1 Introduction As flow passes through a channel constriction most of the energy losses occur as expansion losses downstream of the contraction.10. Stage-discharge relation for a sand channel (Los Gatos Creek at El Dorado Avenue. general scour depths due to constriction. The rise in water level is referred to as the bridge backwater.10 HYDRAULICS OF BRIDGE WATERWAYS 2. Hydraulic engineers are concerned about the computation of backwater with respect to flooding upstream of the bridge.

39. the superstructure must be anchored to counter buoyancy. Backwater in a river with floodplain where the road crossing is more or less normal to the valley but the road approaches block off overbank flow (Figure 2. skewed or curved road embankment as sketched in Figure 2.10. in order to improve the hydraulic efficiency and to reduce the tendency to catch driftwood and ice as illustrated in Figure 2.3 Effects of a Submerged Superstructure If the high-water level reaches the bottom of the superstructure. I through III. The effect of guide banks shown in Figure 2. For bridges which are designed to be submersible under certain conditions.2 Backwater Effects on Waterway Openings It is necessary to distinguish between the following types of backwater effects. The backwater effect arising from this type is seldom large. 2.63 . Backwater in an incised river channel without substantial overbank flow. in supercritical flow water levels may rise suddenly at the contracted section.10. although they might never arise in ordinary bridge design practice. In the case of steep rivers with wide floodplains this effect can be very large. The phenomenon of "choking" is particularly likely if the Froude number only slightly exceeds 1. boundary roughness and slope of the channel must increase from Type I to Type II and to Type III.10. 2. but may be significant in developed areas. "Choking" may occur even in subcritical flow if the constriction is severe enough. in effect.10. In these cases the backwater may be significantly greater than in type b. 2. Wider or additional openings should be designed if choking effects are expected to occur. or by providing dikes (shown on the figure) to close off the affected part of the floodplain from flood waters. Also.4 Effects of Supercritical Flow In contrast to the usual drop at a constriction in subcritical flow. and in part from obstructive effects of piers (Figure 2.38a. This type of effect can be prevented by choosing a suitable location and alignment. The backwater effect along the embankment arises from ponding of water along a line running obliquely down-valley.0. the bridge will act as a short culvert. since a large pond is created. the discharge. where the bridge opening is.38b).38c). located up-valley from one end of the embankment. 2. resulting in part from constriction of flow through an opening somewhat smaller than the natural cross section.38c is to reduce the backwater effects by improving the hydraulic efficiency of the opening.5 Types of Flow in Bridge Openings Three types of flow.40 are often encountered in bridge waterway design.2. it is advisable to provide a rounded nosing on the leading edge of the girder. • • It is advisable to be aware of other unusual backwater effects that might occur in special circumstances. • Backwater on a floodplain resulting from construction of a long. As the scale of the normal depth is the same for all flow profiles. illustrated in Figure 2. or possibly by providing a relief span.

both without and with guide banks (after Neill 1975).Figure 2. (b) effect due to constriction of the channel flow. Three types of backwater effect associated with bridge crossings.38. (c) effect due to constriction of the overbank flow. 2. (a) effect of a skewed embankment across a floodplain.64 .

In Type I flow. the flow is supercritical throughout the reach as the normal water surface is everywhere below critical depth. Referring to Type III flow.14 (SI) and 2. and (3) possible hydraulic jumps near the embankments.Figure 2. nonuniform flow. However.11. However. Army Corps of Engineers (2001) HEC-RAS (River Analysis System) and UNET (Barkau 1993). Solved problems are presented in Section 2.65 . The backwater curve for the water surface elevation upstream from the constriction is independent of the water surface elevation downstream. but not limited to.S. The U.21. An undulating hydraulic jump (with Fr < 2) is formed when the water surface elevation dips below critical depth downstream from the contracted section (Type IIB). significant rise in the water surface might occur in the vicinity of the constriction due to: (1) changes in the specific energy or specific discharge diagram as indicated in Figures 2. In Type II flow. mountainous regions. Backwater should not occur as long as the flow remains supercritical since the flow is controlled from upstream conditions.39.11 COMPUTER MODELS FOR HIGHWAY BRIDGES 2. (2) cross waves and transitions. 2. WSPRO is for steady. 2. Backwater calculations are obtained by applying the conservation of energy principle between Sections 1 and 4. three have the most utility for highway bridge analysis. HEC-RAS is the successor the Corps of Engineers HEC-2 water surface profile program.19 and 2.1 One-Dimensional Computer Models Many one-dimensional computer models are available for computing water surface profiles.15 (English) to illustrate how to calculate maximum constrictions without causing backwater and to calculate water surface elevation upstream of a grade control structure. average depth and velocity for open channel flow. the normal water surface is everywhere above critical depth and the flow is subcritical. Such conditions require steep channels as experienced in. subcritical flow upstream of the bridge passes through critical depth in the constriction. Submergence of a superstructure. These are FHWA's WSPRO (Arneson and Sherman 1998).

66 . Bradley 1978).40.Figure 2. Types of flow encountered (HDS 1. 2.

2.HEC-RAS performs for steady or unsteady. uniform. Army Corps of Engineers HEC-6 (1993) computer programs. and the depth and velocity along and across the stream.12. These are FHWA's BRI-STARS (Molinas 2000) and U. wooded. floodplains into the bridge backwater calculations. at river confluences. uniform or nonuniform flow. 1997. The models are for steady or unsteady and nonuniform open channel flow. They are the FHWA’s FESWMS (Froehlich 1996) and the U. and elliptical. Army Corps of Engineer’s (1997) RMA-2V (Thomas and McAnally 1985) models.2 Two-Dimensional Computer Models Two-dimensional computer models give the water surface profile. Ayres Associates 1994 and 1997. Inlet control occurs when the culvert barrel is capable of carrying more flow than the inlet will accept. can be computed. in conjunction with either free surface or pressure flow through the bridge. and sometimes corrugated plastics. 2.11. They have all the capabilities of the onedimensional programs and have the utility of giving the velocity and depth distribution along and across the channel as a function of time and distance.S. However. 2. corrugated steel. arch and metal box sections are commonly used. pipe arch. Under outlet control conditions. they can give an approximate distribution of the velocity in a cross-section. two have the most utility for highway bridge analysis. In addition. incorporate the effect of wide. Critical flow depth is located at the inlet and the flow is supercritical in the barrel. Embankment overtopping flows. corrugated aluminum.S. and Richardson and Lagasse 1999 pages 701 to 824) Both models require that a grid system be created for the river system. Culvert shapes vary from circular to rectangular. either subcritical or pressure flow exists in the culvert barrel. Of the many models. there are two one-dimensional computer models which include sediment transport. and nonuniform flow. They can and have been used to analyze tidal flows (Zevenbergen et al. The hydraulic design of culverts is given in HDS 5 (FHWA 1985) and HY8 (FHWA 1998) is a computer program for the design and analysis of culvert flows. The programs are capable of computing profiles at stream crossings with multiple openings (including culverts). as well as subcritical and supercritical flow regimes play an important role in determining the location of the control section.1 Introduction A culvert is a conduit which conveys stream flow through a roadway embankment. Two basic types of flow control are recognized depending on the location of the control section: inlet control or outlet control. Outlet control flow occurs when the culvert barrel is not capable of conveying as much flow as the inlet opening will accept. The characterization of pressure. Most culverts are constructed of concrete.12 HYDRAULICS OF CULVERT FLOW 2. UNET is for unsteady. and mixed flow regimes. constructing the grid is greatly aided by the use BYU’s (2000) SMS modeling system.67 . Although one-dimensional. They also.

From Table 2. the following values are obtained: Q A W åVi∆Qi åVi 2∆Qi = = = = = 152.41 cms 138. 2.68 .46 m4/sec2 220.41 indicate how to evaluate the correction factors kt and Cr. The flow over each is calculated separately and the total flow across the roadway is the sum of the incremental flows for each segment (Figure 2.42). A trial and error procedure is necessary to separate the amount of water passing through the culvert.71 m 180.88 m5/sec3 2.1 PROBLEM 1 Evaluation of Correction Factors α and β Calculate the correction factors α and β for a cross-section given the discharge measurement during the peak flood event for the year. from the amount overtopping the roadway.2 or 0.13 ROADWAY OVERTOPPING Roadway overtopping will begin as the headwater rises to the elevation of the lowest point of the roadway. The total flow across the roadway then equals the sum of the roadway overflow plus the culvert flow. This type of flow is similar to flow over a broad crested weir. The flow across the roadway is calculated from the broad crested weir equation Q o = K u k t C r L s (HWr )1. The length of the weir can be taken as the horizontal length across the roadway. If the elevation of the roadway crest varies.0 (English) 9. the vertical curve can be approximated as a series of horizontal segments.20 m2 49.552 (SI) (2.2.5 where: Qo Cr HW r kt Ls Ku Ku = = = = = = = Overtopping discharge in m3/s (ft3/s) Overtopping discharge coefficient Flow depth above the roadway in m (ft) Submergence factor Length of the roadway crest along the roadway in m (ft) 1. Performance curves must then include both culvert flow and road overflow.81 / 32.14 SOLVED PROBLEMS OPEN CHANNEL FLOW (SI) 2.177) The charts in Figure 2. for instance where the crest is defined by a roadway sag vertical curve. if any.4.14.

42. Discharge coefficient for roadway overtopping (after Petersen 1986).41. 2.Figure 2. Weir crest length determinations for roadway overtopping. Figure 2.69 .

682 12..03 11.00 0.44 9.90 m (the maximum depth) y0 3 .564 3.111 15.12 the average values for the energy and momentum coefficients are: and α' = 1.818 0.19 .80 1.0 0.71 138.54 2.732 .22 0.V.44 8.793 2.13 4.. and V.0 0.921 .34 .9 = ≅ 8.442 17.335 m (the smallest non-zero depth) y0 0.95 2.32 1.47 11.44 8.008 2.10 10.236 14.009 1.0 0.335 2.B.723 6.300 0.0 . Richardson.0 TOTAL 49. groundwater and ground response data.60 2.44 8.20 152.0 0.0 0.624 7.882 3.073 13. Civil Engineering Dept. II.93 11.479 3.053 .44 0.411 180.76 0. Stevens.479 3.335 = ≅ 750 0.44 1.15 2.195 .47 2.84 2.27.A.44 8.54 2.00045 ks Using a mean yo/ks of approximately 5.000 0.966 1.000 0.022 . Venezuelan International Meteorological and Hydrological Experiment.024 β' = 1.44 1. August 1971.24 1.409 2.30 6.841 7.023 3.66 2.45 mm and a gradation coefficient G of 3.52 10.32 4.15 1.651 .C.44 8.44 9.32 2.00045 ks and for yo = .137 1. then for yo = 3.14 1.074 0.44 7.800 3.630 3. "Stream flow.365 3. Colorado State University.35 2.0 0.567 1.127 .37 10. E.462 220. M.539 14.99 2.603 12.090 3.026 3.15 1.0 0.169 3.54 2.0 ∆AI (sq m) The bed material at this gaging station has a D50 of 0.731 16.515 0.44 6.000 0.44 9.33 mm and a D65 of 0.88 1 Simons.386 2.165 .192 5.667 22.590 14.917 25.828 26.41 1.692 14. yoi (m) ∆zI (m) VI (mps) 0.23 1.948 12.560 3.49 1.70 .64 12. Vol.44 7.025 2.90 2.22 10.700 0.95 0.88 11.44 8.78 2.703 1.44 8.178 13.68 1.37 2.Table 2.071 16. then from Figure 2.31 1.23 2.245 0.078 3.319 .627 13.282 12.29 2.626 3.4.H.59 2.945 .44 8.78 8. J.93 .0 0. D. Discharge measurement notes1.44 8.297 7.64 10.113 18. Vi2∆QI ∆QI Vi∆QI 4 2 (m /sec ) (m5/sec3) (cms) 0.44 3.101 .16 1. If the value of D65 is used for ks.000. Hydrology Report.096 12.194 14.17 7. Duke.892 7.965 17.64 10. Duke.

270 mm.738 m/sec in a vertical section at a point 244 meters from the right bank. The total discharge is 923. The velocity measurements at various distances from the bottom (y) are shown in Table 2.22 and β = 1.68 0. Equations 2.113 and 2. D65 = 0.024 (138.315 mm and D84 = 0.2 PROBLEM 2 Velocity Profiles and Shear Stress A velocity and discharge measurement is made on the Missouri River at Sioux City.701 0.4) 2 3 = 1.14.762 1.37 0. The average depth is 2.1 should be used for river channels.838 1.As it has been assumed that α' and β' are constant across the river (for convenience).38 meters and the average velocity is 0.094) differ from unity by appreciable amounts.9) (152.4 α = 1. The difference may be important in many river channel calculations.762 0. the assumptions that α = 1.945 2.70 0.152 0.083 These values for α (1.518 0.633 0.991 2.4) 2 = 1.114 become: α = α ′ A 3 å V i2 ∆ Qi 2 Q i and β = β′ A 2 å Vi2 ∆ Qi Q i With the values in Table 2.29 0.42 mm. Table 2.5.000206 and the bed material gradation was determined with D50 = 0.457 0.945 1. If no data are available.884 1.1 cms.305 0.221 β = 1.71 . 2.5. Observed Velocity Data.2) (220. Also during this measurement the slope of the energy grade line was observed as 0.07 0.2) (180.008 (138. y Observed Velocity (m) (m/sec) 0.5) (152.22) and β (1.

17 0.564 (ln y .033 1.75: v/V* = 1/k ln (y/y’) = 2.305 0. v (m/s) v/V* v/V 0.633 9.16 0.945 13.136 1.81 x 2.702 0.962 (b) Power Velocity Distribution Equation The power form of the velocity distribution equation is: v/V = a (y/yo)b.518 7.39 and y' = 0.90 2. The results are: a b R = = = 1.762 1.368 0.70 0.009) y (m) 0.449 0.31/k log (y/y’) Determine the value of k and y': V* = (g yoSf )1/2 = (9.198 1.240 0.457 0.281 1.36 y/y0 0.37 1.577 0.998 2.14 0.38 x 0.ln 0.762 11.07 1.29 Table 2. Velocity Profile Calculations.128 0.281 1.996 giving the values of k = 0.6.838 12.81 0.192 0.321 0.858 0.564 ln (y/0.064 0.950 1.6 is prepared for both log and power forms with the values shown.991 14.069 m/sec Table 2.009. (a) Logarithmic Velocity Distribution Equation The log velocity relation is Equation 2.701 10.945 13.340 0.51 0.705 0.884 12.801 0.152 0. obtain: v/V* = 2.70 0.000206)1/2 = 0.009) with a regression coefficient R = 0.Velocity Profile Analysis The log and power function velocity distribution functions will be used to obtain a mathematical description of the velocity profile.6.72 .04 0. Using regression techniques for the logarithmic form of the equation on the values of v/V* and ln y' with the data in Table 2. To simplify the regression calculations take the ln transform giving: ln v/V = ln a+ b ln (y/yo) using linear regression of ln v/V vs ln (y/yo) calculate the values of a and b.68 1. The final equation for the data is v/V* = 2.

152 ù é ê2.97 τ0 = γ R Sf = 9800 x 2.009 ú ë û 2 = 1000 (0.The resulting power equation is: v/0.75 (v/V* = 2.5.27 y 0 ù é ê5.38 x 0.99 τ0 = ρ v1 − v 2 ( ) 2 2 é y1 ù 5 .100 and the average depth 2.009)) the shear stress on the bed using a single point velocity near the bed is: τ0 = ρv 2 y ù é ê2.4.2) y' = 0.27 m τ0 = ρ V ( ) 2 2 12. average velocity 0.11N / m 2 (c) Local Shear Stress on the Bed Using Two Point Velocities near the Bed For this determination use Equation 2.518 0.738 m/sec. (a) Average Shear Stress on the Bed Calculate the average shear stress on the bed using Equation 2.564 ln 0.738 = 1.564 ln (y/0.305 ù é 5 . Taking ks as (30.564 ln 0. 75 log ê ú y2 ú ê ë û = 1000 0.633 − 0.37 N / m 2 (d) Shear Stress on the Bed Using Average Vertical Velocity Using Equation 2.38 ù é ê5.98 N / m 2 .24 Shear Stress Analysis Using the above information calculate the shear stress on the bed using the several methods given in Section 2.738 ( ) 2 2 12.73 = 3. and a value for ks determine the average shear stress on the bed.37 (y/yo)0.805 N/m2 Using R = yo (for a wide channel) (b) Local Shear Stress on the Bed Using Single Point Velocity near the Bed Using the above information and the logarithmic velocity distribution equation derived from Equation 2.518 ) 2 0.75 log ú 0.38 m.27 x 2.009 ú ë û 2 = 5.75 log ú ks ë û = 1000 0. 75 log ê 0.27 ë û 2.152 ú û ë 2 ( ) 2 = 4.000206 = 4.

2 128.805 newtons per square meter).0020 .0031 . the value from the two velocity method appears to be low.2 150.6 131. This is the average stress on the entire boundary (bed and banks).0001 2. Vi2 ∆ Z = ∆ri (m) i VI (m/s) rI (m) ∆rI (m) gri 0. The detailed calculations based on Equation 2.44 2. very accurate velocity measurements are required.0040 .9 143.7 meters and the outer radius of curvature ro is 156.44 2.0 138. For wide channels the centerline local shear stress on the bed should generally not be lower than the average shear stress on the boundary (i.0000 . Normally the use of the two point velocity equation with the velocities close to the bed is the more accurate.158 are presented in Table 2.921 .31 1.195 . Detailed Computation of Superelevation in Bends.0035 . the value of 4.8 148.1 0.44 2.44 1.3 123.4 meters. however.3 0.44 2.11 in (b). The shear stress computed using velocity is the local shear stress on the bed where the velocity profile was taken.44 2.1 133.44 114.8 155.14.3 PROBLEM 3 Superelevation in Bends Calculate the superelevation of the water surface in a river bend given the velocity profile from Table 2.19 .44 2.732 .44 1.0000 0.(e) Summary The average shear stress on the boundary from (a) is 4. Table 2.0000 1.44 2.7.0014 .0025 .37 in (c).0025 .0001 2.44 111.24 1.805 newtons per square meter.0031 .4 .0004 .44 2.44 116.0015 .0038 .49 1.1 154.5 136.6 .74 .23 1.44 2.0011 2.0026 .6 .14 1.76 118. For this example.300 .0404 m 2.4 140. 4.515 0.e.0031 .966 1. and 3.0024 .44 2..6 153. This local shear stress is calculated as 5.44 2.32 1.7.0030 .9 121.00 Total ∆Z . 2.68 0.44 2.15 1. The river radius of curvature ri is measured equal to 106.82 109.0002 2.0 .00 .00 0.945 .44 2.15 1.41 1.3 145.98 in (d).4.44 2.7 126.165 .61 107.

7) = 0.97 2 ê1 − 12 x 131.05 m.49 m/s (V ) ∆Z = max 2g 2 é ri ê2 − æ ç çr ê è c ê ë ö æ rc ö ù 1.7 ∆Z = 2 ú 2g r c ê 2 x 9. V = Q/A = 1.81 ÷ ÷ ú ø û ù é ú ê ú ê 1 ú = 0.81 ê2 − ç 131.4 − 106. But using the value 0.81 131.7 1.4 ö ú 2 x 9.48 m wide.4 ÷ ÷ ú ê è ø ú û ë Ippen and Drinker’s Equation 2. Neglecting energy losses what is the maximum amount of constriction that a bridge can impose without causing backwater.4 2 ê ë ù ú ú = 0.0718 m è ø è øû ø è oøú ë û 2 2 2 2 The equations give comparable results (0.14.4 ê1 − W 2 ú 12 r c û ê ú ë é ê 1 ê ê 4.∆Z = 1 r0 v2 ∆r å g ri r Total superelevation is 0.4 ) ( ) 2 Ippen and Drinker's Equation 2.10 x 2 x 49.10 ∆Z = V (r 0 − r i ) = (156.10 2 x = ú 2 131.163 é ù ê ú 2 2 2W ê 1 V ú = 1.047 m g rc 9.6 cms and the uniform depth for this discharge is 3.162 é ê 2 ê 2 W 1 ∆Z = V ê 2g r c ê æ W ê1 − ç ç ë è 2rc ù ú ú 2 x 49.042 m ê 2 ê æ 49.4 ö ù ú ê ú ÷ ç ÷ ÷ − ç r ÷ ú = 2 x 9.7 ö ú ê1 − ç ç 2 x 131.041 m ú ú ú û Combined Free and Forced Vortex Equation 2. The design discharge is 141.165 Vmax = 1. the following are obtained: Woodward's Equation 2.0718 m).072 m provides a safety factor.10 m/s. 2. 2.7 ö æ 131.75 .040 to 0. Equation 2.4 ÷ − ç 156.040 meters.4 ÷ ú = 0.81(131.160 2 1.49 2 é æ 106. Based on relationships.4 PROBLEM 4 Maximum Stream Constriction Without Causing Backwater (Neglecting Energy Losses) A stream is rectangular in shape and 30.158 which integrates across the section using the velocity distribution is the most exact.

5 PROBLEM 5 Maximum Water Surface Elevation Upstream of a Grade Control Structure Without Backwater (Neglecting Energy Losses) A low grade control structure (check dam) is to be placed across a stream downstream of a highway bridge. the width of the channel can be contracted until the unit discharge q is a maximum.48 -14. The stream is degrading. From Equation 2.6.The upstream flow rate per unit width (q) is: q= Q 141.05 = 3.17 ÷ ú è3 øú ê ë û 1/ 2 é æ 2 ö 3ù qmax = êg ç H ÷ ú 3 ø ú ê ë è û = 9.6 = 14. The minimum width (maximum constriction) is: W min = Q qmax = 141.17 m 2g 19. when energy losses are considered there will be some backwater at the constriction.62 According to Section 2. Also.76 .2 the maximum unit discharge for a given specific head (specific head equals to a constant) occurs at the critical depth yc where the Froude number is 1. protect the abutments and piers from long term degradation and contraction scour and increase the elevation of the be for 2.8 m This contraction causes the flow to go to critical. This results in an undulating hydraulic jump downstream.149 the critical depth yc for this specific head is: æ q2 ö yc = ç max ÷ ç g ÷ è ø 1/ 3 = 2 2 2 H = Vc 2g 3 1/ 2 3 é æ2 öù = ê9.81 ç x 3.65 cfs / ft W 30.70 = 15.6 = = 1.70 m 9.6 = = 4. cause deposition of bed material.48 The average velocity (V) is: V= Q 141.96 The specific head (H) is: 2 1.63 and the maximum constriction is 30.52 2 H= V + y = + 3.52 m / s A 92.63 m 3 / s / m Therefore. 2. The purpose of the check dam is to maintain the elevation of the water surface at the bridge.14.

At the dam: Hmin + ∆Zmax = 3.local scour. The energy losses are expected to be small so they can be neglected. Sketch of backwater curve over check dam.48 The average velocity (V) is V = Q 141. What is the maximum height of the structure that will not cause backwater at the bridge at the design discharge? The unit discharge (q) in the river at design flood discharge is: q = Q 141. the dam can be built to a height of ∆Zmax which decreases the specific head at the dam to Hmin.43).62 As a first approximation assume no energy loss in the bridge reach.147 Hmin = 2 yc 2. the design flood discharge is 141.05 = 3. Then at the check dam. At the bridge. The river is 30. There is to be no increase in water surface elevation upstream of the bridge at the design discharge.6.6 cms.3 and Equation 2.43.52 m / s A 92.6 = = 4. From Section 2.17 m above the bed (Figure 2.05 m for the design discharge.17 m 2g 19.52 2 H = V + y = + 3. That is.6 = = 1.65 m 3 / s / m W 30. the elevation of the total energy line is 3.77 3 .96 The specific head (H) is 2 1.48 m wide and has a uniform flow depth of 3.17 m Figure 2.

Local scour downstream of the check dam should be evaluated (see HEC-23).93 m 3 / s / m W 30.93 = = 0.22 = 1.61 m / sec y 1.95 = 1. How much backwater will the dam cause for a flow of 28.89 m 3 3 2.45 m = 0.93 2 ù =ê ú ë 9.67 + 1.45 m and from Equation 2.67 m Assuming no energy loss.32 = = 0.52 At the check dam the flow is critical so from Equation 2.65 2 ù = ê ú 9. At lower discharges the check dam will cause backwater.6 cms and the dam will cause no backwater. q = Q 28.22 m If the structure is built to a crest elevation 1.78 .48 The average velocity (V) is V = q 0.6 cms at the check dam will not cause backwater.81 û 1/ 3 = 0.140 é q2 ù yc = ê ú ëgû 1/ 3 é 0.140 é q2 ù = yc ê ú êgû ú ë Then 1/ 3 é 4.17-1.81 ú ê ë û 1/ 3 = 1.30 m Hmin = 2 yc = 2 1.22 m above the bed.95 m and ∆Zmax = 3.144 Hmin = 2 yc = 2 0.From Equation 2.52 m and the dam height is 1. the specific head upstream of the dam is H = Hmin + ∆Z = 0. However.37 m3/s if the normal depth for this discharge is 1.22 m? 3 3 Upstream of the dam the discharge for foot of width (q) is. critical flow will occur at the dam for a flow of 141.30 m = 1. discharges larger than 141.

62 Calculate the correction factors α and β for a cross-section given the discharge measurement during the peak flood event for the year.8 ft (the maximum depth) y0 12.87 .500 ft5/sec3 The bed material at this gaging station has a D50 of 0.27.070 ft4/sec2 85.1 ft (the smallest non-zero depth) y0 1.32 cms. If the value of D65 is used for ks.45 ks 2.22 m high dam when the flow is 28.52 m.1. the following values are obtained: Q A W åVi∆Qi åVi 2∆Qi = = = = = 5.137) H= or y 3 − 1.87 m As the normal depth is only 1. 2.485 ft2 163 ft 21.370 cfs 1.700 0.To determine the depth of flow upstream of the dam (y) solve the specific head equation (Equation 2.79 .45 mm and a gradation coefficient G of 3.8 ) ≅ 8.15 SOLVED PROBLEMS OPEN CHANNEL FLOW (ENGLISH) 2.45 ks and for yo = 1.15.35 m That is. From Table 2.89 m 0.89 y 2 + The solution is y = 1.8 ) ≅ 750 0.8 = (304.1 PROBLEM 1 Evaluation of Correction Factors α and β q2 2gy 2 +y= 0. the depth upstream of the dam is increased 0.1 = (304.35 m by the 1. the backwater is ∆y = 1. then for yo = 12.8.93 2 19.33 mm and a D65 of 0.52 = 0.62 y 2 = 1.93 2 =0 19.

18 715.62 11.95 Vi ∆QI 5 3 (ft /sec ) 0.06 14. J. Stevens.4 78.42 2058.0 0.. Discharge measurement notes1.80 .64 2.65 5530.6 4.98 0.71 4. Vol.00 0.07 376.27 2159.0 8.8 0.45 7251. Civil Engineering Dept.30 4.0 68.5 11.0 8.77 2 5.09 392.74 3.04 163.8 99.0 8.0 8.71 VI∆QI 4 2 (ft /sec ) 0.32 3.0 0.24 6028.62 376.024 and β' = 1.0 8.75 391.H. "Stream flow.00 8.C.68 884.8 12.98 437.96 373.89 3.20 278. D.9 11.8 ∆QI (cfs) 0.0 8.2 92.29 2074.78 3.78 4. then from Figure 2.45 6.0 0.0 8.00 8.0 88. Richardson.V. groundwater and ground response data. E.24 0.5 0.36 392.0 VI (fps) 0.0 8.06 907. 2.64 236.4 84.0 12.0 8.0 8. and V.70 354.8 36.8 9.71 1 Simons.80 6424.0 0.06 3.0 8.19 5357.4 100.27 9.23 23.6 12. yoi (ft) 0.54 0.8 87.8 20.5 8. Equations 2.008 As it has been assumed that α' and β' are constant across the river (for convenience).4 3.0 8.0 0. II.0 8.06 389.43 1463.8.0 8.42 2812.114 become: α = α ′ A 3 å V i2 ∆ Qi 2 Q i Table 2.4 11.0 92. Hydrology Report.0 1484.8 10.03 2803.76 387.0 8.0 1.00 8.0 20.8 91.99 374.4 11.0 TOTAL 163.12 1409.74 21072.02 4.113 and 2.6 11.02 1.69 1686.13 5368.63 4. Venezuelan International Meteorological and Hydrological Experiment.0 ∆zI (ft) 4.6 10.A.Using a mean yo/ks of approximately 5.5 0.44 940.34 1380.86 9696.04 10164.16 423.77 292.8 6.0 102.9 0.0 2.44 216.69 ∆AI (sq ft) 0.94 1417.70 1582.0 8.10 3.12 the average values for the energy and momentum coefficients are: α' = 1.9 5368.0 7. Duke.46 1817.0 11.2 91.4 44.2 86.4 10.90 4.05 127.0 85494. Duke.1 2.77 75.61 5273.0 8.000. August 1971.30 1499.0 8.6 12.87 8416.83 1627. M. Colorado State University.8 96.40 3.28 3.B.2 94..56 7030.17 4.0 8.

070) (5.094) differ from unity by appreciable amounts.25 and β = 1.369 ) 2 3 = 1. y Observed Velocity (ft) (ft/sec) 0. (a) Logarithmic Velocity Distribution Equation 2.70 1.90 5. Observed Velocity Data. and D65 = 0. Iowa.1 should be used for river channels.5 1.008 (1.485 ) (85.25 Also during this measurement the slope of the energy grade line was observed as 0.315 mm.and ' β = β A2 å Vi2 ∆ Qi Q i With the values in Table 2.5 2.50 3. The total discharge is 32.08 1.000206 and the bed material gradation was determined with D50 = 0.600 cfs. The average depth is 7.270 mm.094 These values for α (1.75 4.5 3.25 3.30 2.5 3.8 α = 1.5 2.80 feet and the average velocity is 2.9.10 6. the assumptions that α = 1.5 2.5 2.490 ) (5. If no data are available.369 ) 2 = 1. 2.024 (1. The velocity measurements at various distances from the bottom (y) are shown in Table 2.10 7.81 .42 ft/sec in a vertical section at a point 800 feet from the right bank.2 PROBLEM 2 Velocity Profiles and Shear Stress A velocity and discharge measurement is made on the Missouri River at Sioux City. Table 2. Velocity Profile Analysis The log and power function velocity distribution functions will be used to obtain a mathematical description of the velocity profile.485 ) (21.247 β = 1.0 2.15.247) and β (1.9. The difference may be important in many river channel calculations.

75 12.064 0.281 3. The results are: a b R = = = 1.ln 0.50 1.12 0.50 Table 2.577 0.08 9.50 5.136 2.998 The resulting power equation is: v/2.71 1.60 1.950 2.996 giving the values of k = 0.60 1. To simplify the regression calculations take the ln transform giving: ln v/V = ln a+ b ln (y/yo) using linear regression of ln v/V vs ln (y/yo) calculate the values of a and b.128 0.75: v/V* = 1/k ln (y/y’) = 2.340 y/y0 0. Velocity Profile Calculations.42 = 1.96 1.564 ln (y/0.192 0.198 3.2 x 7.50 3.03.00 1.10 13.25 14.801 0.321 0.03) with a regression coefficient R = 0.10 obtain: v/V* = 2.50 6.240 0.70 7.80 x 0.564 (ln y .38 (y/yo)0.281 3.705 0.39 and y' = 0. for the logarithmic form of the equation on the values of V/V* and ln y' with the data in Table 2.702 2.30 10.50 10.10.03) y (ft) 0.The log velocity relation is Equation 2.10 13.10 is prepared for both log and power forms with the values shown.06 1.000206)1/2 = 0.25 1.25 7.860 2.90 12.50 4.449 0.82 . Using regression techniques. v (ft/s) v/V* v/V 1.46 0.962 (b) Power Velocity Distribution Equation The power form of the velocity distribution equation is: v/V = a (y/yo)b.50 2. The final equation for the data is v/V* = 2.228 ft/sec Table 2.31/k log (y/y’) Determine the value of k and y': V* = (g yoSf )1/2 = (32.24 2.033 2.368 0.09 0.

8 ù é ê5.70 ) 2 0.03)) the shear stress on the bed using a single point velocity near the bed is: τ0 = ρv 2 y ù é 2 .100 and the average depth 7.4 x 7.27 x 7. (a) Average Shear Stress on the Bed Calculate the average shear stress on the bed using Equation 2.75 log 0. 564 ln ê 0.5 ú ë û 2 ( ) = 0.5.084 lb / ft 2 2.03 ú ë û 2 = 0.4.70 1 . 564 ln ê 0. and a value for ks determine the average shear stress on the bed.8 x 0. 75 log ê 0 .Shear Stress Analysis Using the above information calculate the shear stress on the bed using the several methods given in Section 2.75 (v/V* = 2.99 τ0 = ρ v1 − v2 ( ) 2 2 y ù é ê5.75 log ú ks û ë = 1. average velocity 2.564 ln (y/0.108 lb / ft 2 (c) Local Shear Stress on the Bed Using Two Point Velocities near the Bed For this determination use Equation 2.08 − 1.100 lb/ft2 Using R = yo (for a wide channel) (b) Local Shear Stress on the Bed Using Single Point Velocity near the Bed Using the above information and the logarithmic velocity distribution equation derived from Equation 2.75 log 1 ú y2 ú ê ë û = 1.42 ft/sec.000206 = 0.2) y' = 0.094 lb / ft 2 2 (d) Shear Stress on the Bed Using Average Vertical Velocity Using Equation 2.91 ú ë û = 0.5 ù é 2 .27 y o ù é ê5.94 2.42 ( ) 2 2 12. Taking ks as (30.91 τ0 = ρ (V ) 2 2 12.83 .0 ù é 5 .94 2.97 τ0 = γ R Sf = 62.94 (1.03 ú ë û 2 = 1.80 ft.

0046 .0 8.0115 .40 .74 3. For wide channels for the centerline local shear stress on the bed should generally not be lower than the average shear stress on the boundary.0 8.0 8.64 .0003 8. For this example.0079 . Table 2.06 3.89 3.0126 . The shear stress computed using velocity is the local shear stress on the bed where the velocity profile was taken.11.0101 .108 in (b).0 8. the value of 0. Normally the use of the two point velocity equation with the velocities close to the bed is the more accurate. Vi2 ∆ Z = ∆ri (ft) i VI (ft/s) rI (ft) ∆rI (ft) gri 2.78 4.0064 .0049 .0 8.0 382 2.0000 6.0082 .69 0.158 are presented in Table 2.0 374 0.0101 ..00 0. the value from the two velocity method appears to be low.78 3.63 4. however.00 Total ∆Z .71 4.0 352 0. 0.e.0 366 0. in (d).02 4.84 .54 .0 358 0.5 390 398 406 414 422 430 438 446 454 462 470 473 486 494 502 507 512 3.1330 ft 2. The detailed calculations based on Equation 2.15.10 pound per square foot. 0.0 5.5 2.0000 .0 8. This local shear stress is calculated as 0.0005 8.0 8.30 4.0131 . The river inner radius of curvature ri is measured equal to 350 feet and the outer radius of curvature ro is 513 feet. This is the average stress on the entire boundary (bed and banks).0 8.0086 .8. very accurate velocity measurements are required.0099 .98 .0002 8.17 4.32 3.0000 0.084.0 8.11. i.00 .0037 8.0 8.(e) Summary The average shear stress on the boundary from (a) is 0.0083 . Detailed Computation of Superelevation in Bends.3 PROBLEM 3 Superelevation in Bends Calculate the superelevation of the water surface in a river bend given the velocity profile from Table 2.02 1.0103 .0 7.10 3.0014 . 2.0 8.094 in (c).10 pounds per square foot.90 4.0 8.0 8.

236 ft ú 2 x 32.151 ft 2 ê ú ê1 − 163 2 ú 12 x 431 û ê ú ë Combined Free and Forced Vortex Equation 2.2 431 ê1 − W 2 ú 12 r c ú ê ë û 2 é ù ê ú 1 ê ú = 0. Equation 2.133 to 0. The design discharge is 5.∆Z = 1 r0 v2 ∆r å g ri r Total superelevation is 0.165 Vmax = 4.61 ft/s.90 ê2 − æ ç ÷ −ç ÷ ú = 0. Neglecting energy losses what is the maximum amount of constriction that a bridge can impose without causing backwater. 2.158 ft ê 2 ê æ 163 ö ú ê1 − ç ç 2 x 431 ÷ ÷ ú ê è ø ú û ë Ippen and Drinker’s Equation 2. But using the value 0.24 ft provides a safety factor.4 PROBLEM 4 Maximum Stream Constriction Without Causing Backwater (Neglecting Energy Losses) A stream is rectangular in shape and 100 ft wide.85 .2 ê è 431 ø è 513 ø ú ú ë û û The equations give comparable results (0.2 (431) ( ) 2 Ippen and Drinker's Equation 2.2 ú û ù é ú ê ú ê 1 ú = 0.160 2 3.162 é ê 2 2W ê 1 V ∆Z = ê 2g r c ê æ W ê1 − ç ç ë è 2rc ö ÷ ÷ ø 2 ù ú ú 2 x 163 3.61 ∆Z = V (r 0 − r i ) = (513 − 350 ) = 0.236 ft).000 cfs and the uniform depth for this discharge is 10 ft.90 ft/s (V ) ∆Z = max 2 2g é ri ê2 − æ ç ç ê è rc ê ë ö æ rc ö ÷ ÷ −ç ç ÷ ÷ ø è ro ø 2 2 2 2 ù 2 é 350 ö æ 431 ö ù ú = 4.163 é ù ê ú 2 2W ê 1 V ú = 3.153 ft g rc 32.133 feet. the following are obtained: Woodward's Equation 2.61 x 2 x 163 ∆Z = 2 ú 2g r c ê 2 x 32.15. V = Q/A = 3. Based on relationships. 2.158. which integrates across the section using the velocity distribution is the most exact.612 x = ú 431 ú 2 x 32.

39 ft 2g 64.000 = = 50 cfs / ft W 100 The average velocity (V) is: V= Q 5. 2.2 ç x 10.39 ÷ ú ø ú è3 ê ë û 1/ 2 = 103.00 ft / s A 1.4 cfs / ft Therefore.86 .2 the maximum unit discharge for a given specific head (specific head equals to a constant) occurs at the critical depth yc where the Froude number is 1.000 = = 5.4 and the maximum constriction is 100 – 48.The upstream flow rate per unit width (q) is: q= Q 5. the width of the channel can be contracted until the unit discharge q is a maximum.3 = 51.6.5 PROBLEM 5 Maximum Elevation of a Grade Control Structure Without Backwater (Neglecting Energy Losses) A low grade control structure (check dam) is to be placed across a stream downstream of a highway bridge. The stream is degrading.000 The specific head (H) is: H = V + y = 5 + 10 = 10. when energy losses are considered there will be some backwater at the constriction. This results in an undulating hydraulic jump downstream. The minimum width (maximum constriction) is: W min = Q qmax = 5.149 the critical depth yc for this specific head is: 2 2 æ q2 yc = ç max ç g è ö ÷ ÷ ø 1/ 3 = 2 2 2 H = Vc 2g 3 Solving for qm qmax é æ 2 ö3 ù = êg ç H ÷ ú ê ë è3 ø ú û 1/ 2 3 é æ2 ö ù = ê32. Equation 2.15.4 According to Section 2. The purpose of the check dam is to maintain the 2.000 = 48.7 ft This contraction causes the flow to go to critical. Also.3 ft 103.

000 The specific head (H) is V = H = V + y = 5 + 10 = 10. Sketch of backwater curve over check dam.44.44).elevation of the water surface at the bridge. There is to be no increase in water surface elevation upstream of the bridge at the design discharge. At the bridge.39 ft 2g 64.00 ft / sec A 1. cause deposition of bed material.39 ft above the bed (Figure 2. The river is 100 ft wide and has a uniform flow depth of 10 ft for the design discharge. the design flood discharge is 5000 cfs. What is the maximum height of the structure that will not cause backwater at the bridge at the design discharge? The unit discharge (q) in the river at design flood discharge is: q = Q 5. 2.4 As a first approximation assume no energy loss in the bridge reach.87 .39 ft 2 2 Figure 2. The energy losses are expected to be small so they can be neglected. the elevation of the total energy line is 10. Then at the check dam.000 = = 50 cfs / ft W 100 The average velocity (V) is Q 5.000 = = 5. At the dam: Hmin + ∆Zmax = 10. protect the abutments and piers from long term degradation and contraction scour and increase the elevation of the be for local scour.

How much backwater will the dam cause for a flow of 1000 cfs if the normal depth for this discharge is 5 ft and the dam height is 4. Local scour downstream of the check dam should be evaluated.000 cfs check dam will not cause backwater.6.000 cfs and larger and the dam will cause no backwater.88 .0 ft above the bed.27 ft Hmin = 2 yc = 2 4.2 û 1/ 3 = 4. the dam can be built to a height of ∆Zmax which decreases the specific head at the dam to Hmin.40 = 4.39 .000 = = 10 cfs / ft W 100 The average velocity (V) is V = q 10 = = 2. HEC-23 has am problem for determining this scour.00 ft / sec y 5 At the check dam the flow is critical so from Equation 2. From Section 2.27 ft = 6.6. critical flow will occur at the dam for a flow of 5.0 ft If the structure is built to a crest elevation 4. q = Q 1.3 and Equation 2.0 ft? 3 3 Upstream of the dam the discharge for foot of width (q) is.That is.2 û 1/ 3 = 1. discharges larger than 5.140 3 é q2 ù yc = ê ú êgû ú ë Then 1/ 3 é 50 2 ù = ê ú ë 32. At lower discharges the at the check dam will cause backwater.142 é q2 ù yc = ê ú ê ëgú û 1/ 3 é 10 2 ù = ê ú ë 32.46 ft 2.40 ft and ∆Zmax = 10.147 Hmin = 2 yc From Equation 2. However.

137) H = or y3 .19 y2 + 102/64.14 ft by the 4.4 = 0 The solution is y = 6.19 ft To determine the depth of flow upstream of the dam (y) solve the specific head equation (Equation 2.000 cfs.4 y 2 3 3 2.46 ft = 2. the specific head upstream of the dam is H = Hmin + ∆Z = 2.19 + 4. the backwater is ∆y = 6. 2 q 10 + y = 6.144 Hmin = 2 yc = 2 1.6. the depth upstream of the dam is increased 1.and from Equation 2.14 ft That is.5.14 ft As the normal depth is only 5 ft.19ft + y = 2 2 2g y 64.89 .0 = 6.19 ft Assuming no energy loss.0 ft high dam when the flow is 1.14 .00 = 1.

(page intentionally left blank) 2.90 .

and can still be.1 . size represents a sufficiently complete description of the sediment particle for many practical purposes.004 mm) forming a binder with silts and sand. That is. bed configuration and resistance to flow are a function of the flow and can change to increase or decrease the velocity and water surface level. This chapter presents the fundamentals of alluvial channel flow. The non-cohesive material generally consists of silt. or cobbles. methods of measuring properties of alluvial materials. prediction of bed forms. In alluvial rivers.S. Manning's n for sandbed and other natural streams. transported by the stream. Therefore. sediment transport and physical and computer modeling of sediment transport will be covered. The river channel can shift its location so that the crossing or encroachment is unfavorably located with respect to the direction of flow.CHAPTER 3 FUNDAMENTALS OF ALLUVIAL CHANNEL FLOW 3. beginning of motion. The particle size is the most readily measured property. Because of the electro-chemical bonding between clay particles. flow in sandbed channels. 1957. These fundamentals of alluvial channel flow are used in later chapters to develop design considerations for highway crossings and encroachments in river environments. Several of the important bed material properties are discussed in the following sections (U. Richardson 1971). It covers properties of alluvial material. physical size has by far the greatest significance to the hydraulic engineer. 3.2. as they indicate the behavior of the particles in their interaction with the flow. In Chapter 4. The moveable boundary of the alluvial river adds another dimension to the design problem and can compound environmental concerns. Silt generally is not present in appreciable quantities in streams having non-cohesive boundaries.1 INTRODUCTION Most streams that a highway will cross or encroach upon are alluvial. the rivers are formed in cohesive or non-cohesive materials that have been.2 SEDIMENT PROPERTIES AND MEASUREMENT TECHNIQUES A knowledge of the properties of the bed material particles is essential. and other properties such as shape. Cohesive material consists of clays (sizes less than 0. or any combination of these sizes.1 Particle Size Of the various sediment properties. sand. 3. clays are more resistant to erosion than silts. how bed-form changes affect highways in the river environment. Interagency Subcommittee 1941. 3. and physical measurement and determination of sediment discharge in the field. gravel. In general. fall velocity and specific gravity tend to vary with size in a roughly predictable manner. flow in coarse-material streams. 1943. the design of highway crossings and encroachments in the river environment requires knowledge of the mechanics of alluvial channel flow.

because of simplicity and effectiveness of correlation with the behavior of particles in flow. As an approximation.062 mm). silts.The diameter of a sphere equal to the length of the side of a square sieve opening through which measured quantities (by weight) of the sample will pass. However. sand (0. gravels.0 . these definitions also depend on the shape and density of the particle. and 4c are the dimensions of the three mutually perpendicular axes of a particle 4a the longest.The diameter of a sphere with the same fall velocity and specific gravity as the particle in the same fluid under the same conditions. distilled water of infinite extent and at a temperature of 24°C. The silt and clay class is of considerable importance in the evaluation of stream sediment loads.65 and also has the same terminal settling velocity as the particle when each is allowed to settle alone in quiescent. Roundness is defined as the ratio of the average radius of curvature of the corners and edges of a particle to the radius of a circle inscribed in the maximum projected area of the particle. The cobble and gravel class plays a considerable role in the problems of local scour and resistance to flow and to a lesser extent in bed load transport. 2. sediments have been classified into boulders. diameter.1. 4. and 4c the shortest axis. The size range in each general class is given in Table 3. In general.The diameter of a sphere that has a specific gravity of 2.0 mm). Sphericity is defined as the ratio of the surface area of a sphere of the same volume as the particle to the actual surface area of the particle. weight. 3.S.1) where: 4a.2. 4b the intermediate.2 Particle Shape Generally speaking. cobbles.64 mm). The boulder class (250 . 3.Particle size Ds may be defined by its volume. 3. Sp. Sieve diameter . the most commonly used parameter to describe particle shape is the Corey shape factor. (Albertson 1953) defined as: Sp = lc l al b (3. gravel (2.4000 mm) is generally of little interest in sediment problems. The sand class is one of the most important in alluvial channel flow. 4b. Sedimentation diameter . The non-cohesive material generally consists of silt (0. Standard fall diameter . The following definitions are commonly used to describe the particle size (U. the sieve diameter is equal to the nominal diameter.2 .004-0. or cobbles (64-250 mm). Except for volume.The diameter of a sphere having the same volume as the particle. Nominal diameter . or sieve mesh size. shape refers to the overall geometrical form of a particle. bank stability and problems of seepage and consolidation. Interagency Subcommittee 1943): 1. sands.2. fall velocity. and clays on the basis of their nominal or sieve diameters.062 .

6 0.031-0.125 01.008 0.25-0.00 1.004 0.3-0. Sediment Grade Scale (Brown 1950).0005 0.50 0.3-0.008-0.062 0.00-1.S.25-0.0020-0.0020 0.25 0.08 ----------------------------------------------------- Very large boulders Large boulders Medium boulders Small boulders Large cobbles Small cobbles Very coarse gravel Coarse gravel Medium gravel Fine gravel Very fine gravel Very coarse sand Coarse sand Medium sand Fine sand Very fine sand Coarse silt Medium silt Fine silt Very fine silt Coarse clay Medium clay Fine clay Very fine clay 3.031 0. Size Approximate Sieve Mesh Openings Per Inch U.016-0.0002 Microns --------------------------------------------2000-1000 1000-500 500-250 250-125 125-62 62-31 31-16 16-8 8-4 4-2 2-1 1-0.24 Inches 160-80 80-40 40-20 20-10 10-5 5-2.16 0. Tyler Standard --------------------------------2 1/2 5 9 16 32 60 115 250 ------------------------------------5 10 18 35 60 120 230 Class Millimeters 4000-2000 2000-1000 1000-500 500-250 250-130 130-64 64-32 32-16 16-8 8-4 4-2 2-1 1-1/2 1/2-1/4 1/4-1/8 1/8-1/16 1/16-1/32 1/32-1/64 1/64-1/128 1/128-1/256 1/256-1/512 1/512-1/1024 1/1024-1/2048 1/2048-1/4096 --------------------------------------------2.0010-0.0010 0.00-0.5-1.3 .6-0.3 0.Table 3.0005-0.5 2.16-0.016 0.004-0.5-0.062-0.3 1.5 0.1.50-0.

FD = Ws (ρ s − ρ) gVp = C D A s ρ ω 2 / 2 (3. or a pressure difference on the particle or a combination of the two forces. and the water is not distilled or quiescent. If the particle is falling at its terminal velocity. little is known about the effect of turbulence on fall velocity. However. The fluid drag on the falling particle FD is given by the drag equation FD = CD A s ρ ω 2 / 2 (3. A particle falling at terminal velocity in a fluid is under the action of a driving force due to its buoyant weight and a resisting force due to the fluid drag.3.2) The buoyant weight of the particle W s is: Ws = (ρ s − ρ) gVp (3.3) where: CD ω As ρ ρs g Vp = = = = = = = Coefficient of drag Terminal fall velocity of the particle Projected area of the particle normal to the direction of flow Fluid density Particle density Acceleration due to gravity Volume of the particle The area and volume can be written in terms of the characteristic diameter of the particle Ds or: A s = K1 D2 s (3.4 .3 Fall Velocity The prime indicator of the interaction of sediments in suspension within the flow is the fall velocity of sediment particles. K1 = π/4 and K2 = π/6 for spherical particles. Measurement techniques are available for determining the fall velocity of groups of particles in a finite field in fluid other than distilled water. Fluid drag is the result of either the tangential shear stress on the surface of the particle. The fall velocity of a particle is defined as the velocity of that particle falling alone in quiescent.6) 3. distilled water of infinite extent.4) and Vp = K 2 D 3 s (3. For example. the particle is not falling alone. In most cases.5) Where the coefficients K1 and K2 depend on the shape of sediment particles.2.

Drag coefficient CD vs particle Reynolds number Rep for spheres and natural sediments with shape factors Sp equal to 0.1. the shape. fall velocity ω and temperature To (Brown 1950. The relations between the fall velocity of particles and the other variables are given in Figures 3.5 into Equation 3. .8) The drag coefficient CD is dependent on the particle Reynolds number (Re = ρ ωDs/µ).1 and 3.By substituting Equations 3. Also.7) is obtained.5 . p.9. 781). 3. where µ is dynamic viscosity of the fluid. and the surface texture of the particle.3. Four dimensionless variables describing fall velocity result from dimensional analysis of Equation 3. 0. D ç gD ÷ ρ K s 1 è ø (3.7.1).5. Figure 3. sediment diameter Ds vs.4 and 3. the expression: ω2 = ö 2 K 2 æ ρs ç − 1÷ ç ÷ gD s CD K 1 è ρ ø (3. C . The ratio K2/K1 is usually replaced by the Corey shape factor Sp (Equation 3.6. and 0.2.7: æ ω2 ρs K 2 ö ÷=0 fç . 0.

visual accumulation (VA) tube analysis.S.062-2 mm ----0.000-5. the sediment should be wet-sieved using distilled water and a 250-mesh (0.05-15. All together the four methods provide a means of obtaining particle size distributions for most bed material samples. Nominal diameter vs.6 . and pipette analysis.2.2.2).062 mm 2.0 Pipette 0. Each method for size distribution analysis is appropriate for only a particular range of particle sizes (Table 3. 3. fall velocity (Temperature = 24°C) (Guy 1977. Interagency 1957b).062-32 mm ----100-500 VA tube 0.Figure 3.000 1.0-5.062 mm) it must be separated prior to analysis.4 Sediment Size Distribution Four methods of obtaining sediment size distribution are described below: Sieve analysis. U. The material passing through the sieve can be analyzed by pipette analysis if further breakdown of the fine sediment is desired. or dried and included 3.2. pebble count method.062 mm) sieve. Table 3. To separate the coarser from the finer sediment.002-0. Guide to Size Range for Different Types of Size Analysis. Analysis Quantity of Size Concentration Sediment (g) Range (mg/l) or Pebbles Sieves 0.0 Pebble Count 12-1000 mm ----100 pebbles If the sediment sample to be analyzed (bed material or suspended sediment) has considerable fine material (Ds < .

The equipment for the visual accumulation tube analysis consists of: (1) a glass funnel about 250 mm (9. Sieve Size Distribution in the sand and gravel range is generally determined by passing the sample through a series of sieves of mesh size ranging from 32 mm to 0. Then. It is a fast.0 mm). The large material sizes are measured in situ by laying out a square grid. Within the grid. Pebble Count Method is used to obtain the size distribution of coarse bed materials (gravel and cobbles) which are too large to be sieved.062 mm < Ds < 2.0 mm or larger. all the particle sizes are measured and counted by size intervals.8 in. Visual Accumulation (VA) Tube is used for determining the size distribution of the sand fraction of sediment samples (0. If the material is to be analyzed by wet-sieving or with the accumulation tube. Visual accumulation tube method may not be suitable for some streams that transport large quantities of organic materials such as root fibers. economical. (3) glass sedimentation tubes having different sized collectors. Very often the coarser material is underlain by sands. and accurate means of determining the fall velocity or fall diameter of the sediment. with a special clamping mechanism serving as a "quick acting" valve.062 mm.062 mm with the analysis of the coarser material. leaf fragments. (4) a tapping mechanism that strikes against the glass tube and helps keep the accumulation of sediment uniformly packed. It has been shown that particles of a sample in the visual tube settle with greater velocities than the same particles falling individually because of the effect of mutual interaction of the particles. the underlying sands are analyzed by sieving. A minimum of about 100 grams of sand is required for an accurate sieve analysis. extra care is needed when a stream transports large quantities of heavy or light minerals such as taconite or coal. For large samples a random selection of particles in the various classes is appropriate to develop frequency histograms of sediment sizes. If it is going to be dry-sieved. and (6) the recorder chart which has a printed form incorporating the fall diameter calibration. The two classes of bed material are either combined into a single distribution or used separately. Standard methods employed in soil mechanics are suitable for determining the sieve sizes of sand and gravel sediment samples.) long. 3. and algae. (2) a rubber tube connecting the funnel and the main sedimentation tube. it is not dried. the particles start falling from a common source and become stratified according to settling velocities. the particles coming to rest at the bottom of the tube are of one "sedimentation size" and are finer than particles that have previously settled out and are coarser than those remaining in suspension. Also. In the visual accumulation tube method.7 .as percent finer than 0. At a given instant. The method is explained in detail by Guy (1977). More is required if the sample contains particles of 1. The visual accumulation tube apparatus is calibrated to account for the effects of this mutual interaction and the final results are given in terms of the standard fall diameter of the particles. the material retained on the sieve is oven dried for one hour after all visible water has been evaporated. (5) a special recorder consisting of a cylinder carrying a chart that rotates at a constant rate and a carriage that can be moved vertically by hand on which is mounted a recording pen and an optical instrument for tracking the accumulation.

The presentation of sediment size analysis is made in various formats. is imaged by a lens onto the plane of a Plexiglass plate. If the class intervals are small. An example of a histogram is shown in Figure 3. As the different diameters are registered. The successive sizes employed in the size analysis of sediment are usually in ratios of 2 or 2 .6 ft) above the stream bed. A reference scale. special apparatus can be set up for the analysis of a large number of samples. By adjusting the iris diaphragm the diameter of the sharply defined circular light spot appearing on the photograph can be changed and its area made equal to that of the individual particles.Square-Surface Sample is obtained by picking up and counting all the surface pebbles in each predetermined size class within a small enclosed area of the bed.8 . such as a steel tape or a surveyor's rod must appear in the photograph. the shape of the histogram will approach a continuous curve. and the ordinate scale represents either actual concentration or percent (by number. 3. An efficient operation can count up to 1. The time and depth of withdrawal are predetermined on the basis of Stokes law. usually in geometric progression. The photographs are printed on the thinnest paper available. In the line sampling method of pebble count sampling. Particles having a settling velocity greater than that of the size at which separation is desired will settle below the point of withdrawal after elapse of a certain time. or weight) of the total sample contained in each class interval. volume. The upper size limit of sediment particles which settle in water according to Stokes is about 1/16 mm or 0. The area is taken to be representative of the whole channel bed. The fundamental principle of the pipette method is to determine the concentration of a suspension in samples withdrawn from a predetermined depth as a function of settling time. for routine analysis. illuminated from one side. An iris diaphragm. For this method. or volume percentage of items in given class intervals. a line is laid out or placed either across or along the stream. A histogram is a graphical representation of the number. Therefore. use of the Zeiss Particle-Size Analyzer should be considered (Ritter and Helley 1968). with a 35 mm camera supported by a tripod about 2 m (6. The pebble count method entails measurement of randomly selected particles in the field. A complete description of a laboratory setup and procedure for this method is given by Guy (1969). a puncher marks the counted particle on the photograph. Usually 100 particles are sufficient to give an accurate classification of the size distribution of coarse materials. often under difficult conditions. The abscissa scale represents the class intervals. a photograph of the stream bed is made.1) and is an important division in many phases of sediment phenomena. weight. Satisfactory use of the pipette method requires careful and precise operation to obtain maximum accuracy in each step of the procedure.062 mm. This size is the division between sand and silt (Table 3. Frequency Curves.3a. the height depending on the size of the bed material. This corresponds to the lower size limit which can be determined readily by sieves.062 mm is one of the most widely accepted techniques utilizing the Oden theory and the dispersed system of sedimentation. The measured particles are classified as to size or weight and a percent finer curve or table is prepared. Pipette Method of determining gradation of sizes finer than 0. Particles are picked at random intervals along the line and measured.000 particles in 30 minutes. Also. preferably at low flow.

Definition sketches for size-frequency characteristics of sediments (Rouse 1940).3.9 .Figure 3. 3.

10) for the effective or representative grain size of well-graded materials. Measures of spread are based on differences or ratios between the two quartiles. D85. it is possible to choose certain particle sizes as representing significant values. In a size frequency distribution curve. and most of the data may be obtained directly from the cumulative curve by graphic means.3). In studies of scour below culvert outlets.9) where: G = Dx = Gradation coefficient Sediment diameter particle of which x percent of sample is finer The grain roughness used in velocity equations is taken as the D80. Commonly. If this is true. the size distribution of natural sediments plots as a straight line on log probability paper. Quartile measures are very readily computed. D85. It is computed with the expression G= 1 é D 50 D 84 ù + ê ú 2 ë D16 D 50 û (3. Di. In contrast to quartile measures. Here 3. The slope of this line is proportional to the spread of the size distribution in a sediment sample. sizes can be determined. and particles just larger than three-fourths of the distribution D75 (the third quartile). In this diagram. moment measures are influenced by each individual size class in the distribution. etc. or D90.10 . such as particles just larger than one-fourth of the distribution D25 (the first quartile). the cumulative percent finer distribution diagram is obtained (Figure 3. then a natural sediment is completely described by the median diameter (D50 the size of sediment of which 50 percent is finer) and the slope of the cumulative frequency line on log probability paper. ì 10 3 ü å Di ï ï ï ï k = í i=1 ý ï 10 ï ï ï þ î 1/ 3 (3. From the cumulative percent finer the D35. Stevens (1968) was able to consolidate a wide range of scour data by employing the expression. the abscissa scale (usually logarithmic) represents the intervals of the size scale and the ordinate scale is the cumulative percent by weight of the sample up to (or percent finer than) the size in question. Quartile measures are confined to the central half of the frequency distribution and the values obtained are not influenced by larger or smaller sizes.When the ordinates of successive classes are added and plotted against the upper limit of the size class. The second moment is a measure of the average spread of the curve and is expressed as the standard deviation of the distribution. D50. The first moment of a frequency curve is its center of gravity and is called the arithmetic mean and is the average size of the sediment.

5 Specific Weight Specific weight is weight per unit volume. To determine porosity. 3. D10. in grams per cubic centimeter. are the sieve diameters of the riprap for which 0 percent. Next. Stevens’ equation is the equivalent to utilizing the arithmetic average of the sum of the weights of the individual particles. the specific weight is the weight of solids per unit volume of the material including its voids. or water sediment mixtures.12) Where pI is the percentage by weight of that fraction of the sediment with geometric mean size Dsi. 10 percent.6 Porosity The porosity of granular materials is the ratio of the volume of void space to the total volume of an undisturbed sample. sediment deposits. The effective diameter Dm is also used in sediment studies. The geometric mean size is the square root of the product of the end points of a given size range. 3.2. 100 percent of the material (by weight) is finer.11 . It is defined by the following equation: n p D å D m = i=1 i si 100 (3.11) D i (i = 2) = D i (i = 10 ) = The terms D0. In connection with granular materials such as soils.2. D100. The measurement of the specific weight of sediment deposits is determined simply by measuring the dry weight of a known volume of the undisturbed material. 3. …. The void volume is then obtained by subtracting the volume of solids from the total volume. …. the volume of the sample must be obtained in an undisturbed condition.D i (i = 1) = D 0 + D10 2 D10 + D 20 2 D 90 + D100 2 (3. the volume of solids is determined either by liquid displacement or indirectly from the weight of the sample and the specific gravity of material. In the English system of units. specific weight is usually expressed in units of pounds per cubic foot and in the metric system.

cohesive soils are not classified by grain size but by their degree of plasticity. and in the preservation and enhancement of the stream environment. Simons developed Figure 3.4 for the angle of repose for dumped granular material. 3. the sand material is easily eroded and is continually being moved and shaped by the flow. ionic valence. is related to shear stress. silt and clay soils have more than 50 percent by weight of particles passing the 0. The EFA develops an erosion rate-shear stress curve for a soil sample taken with an ASTM standard Shelby tube with a 3. In sandbed rivers. Briaud et al. given in N/m2. In the unified soil classification system.3 FLOW IN SANDBED CHANNELS Most larger streams flow on sandbeds for the greater part of their length. Cohesive soils are composed of silts and clay. there are potentially many more opportunities for highway crossings or encroachments on sandbed streams than in cohesive or gravel streams. imposed at the soil water interface using an erosion function apparatus (EFA). which is measured by Atterberg limits. It is a measure of the intergranular friction of the material. held together by strong hydrogen bonds. (1999) describe equipment and methods developed to determine the erosion rate of cohesive soils.3. The two main types of clays are kaolinite and montmorillonite.8 Angle of Repose The angle of repose is the maximum slope angle upon which non-cohesive material will reside without moving. Kaolinite crystals are large (70 to 100 layers thick).2. the vane shear test. The mobility of the sandbed creates problems for the safety of any structure placed in or over the stream.2) mm outside diameter. and hydration and swelling properties of the constituent minerals. and are not readily dispersible in water. Clays are alumino-silicate crystals composed of two basic building sheets. Erosion rate is defined as the vertical distance scoured per unit of time.12 .7 Cohesion Cohesion is the force by which particles of clay are bound together. which is performed in the field is one of the simplest. for the protection of private property along these streams. Montmorillonite crystals are small (3 layers thick) held together by weak bonds between adjacent oxygen layers and are readily dispersible in water into extremely small particles. The vane is forced into the ground and then the torque required to rotate the vane is measured. or shear strength.1.0 inches (76. Among these. the tetrahedral silicate sheet and the octahedral hydrous aluminum oxide sheet. Various types of clays result from different configurations of these sheets. 3. Several laboratory and field measurement techniques are available for determining the magnitude of cohesion. particle spacing. 3. expressed as mm/hr. Their size classification is given in Table 3. The shear strength is determined from the torque required to shear the soil along the vertical and horizontal edges of the vane. Thus. The erosion rate. of clays.075 mm sieve opening.2. salt concentration in the fluid. This force is the result of ionic attraction among individual particles. and is a function of the type of mineral. The erosion rate-shear stress curve for samples taken at a site can then be used to determine the scour depth as a function of time. However.

The interaction between the flow and bed material and the interdependency among the variables makes the analysis of flow in alluvial sandbed streams extremely complex. and (3) dunes The bed roughness ranges from dunes to plane bed or antidunes. change with different bed configurations. However. (1) plane bed. process of energy dissipation. and phase relation between the bed and water surfaces. 3. b) with breaking antidunes. The two regimes and their associated bed configurations shown in Figure 3.1 Regimes of Flow in Alluvial Channels The flow in alluvial channels is divided into lower and upper flow regimes separated by a transition zone (Simons and Richardson 1963.5 are: Lower flow regime: Transitional Flow Regime: Upper Flow Regime: (1) ripples. 1966).13 . In the extreme case. mode of sediment transport.4. These two flow regimes are characterized by similarities in the shape of the bed configuration. a three-fold increase in Manning's n results in a doubling of the flow depth. a) with standing waves. 3. The gross measures of channel flow. and (3) chutes and pools. For a given discharge and channel width.Figure 3. with an understanding of the different types of bed forms that may occur and a knowledge of the resistance to flow and sediment transport associated with each bed form. bed elevation and flow velocity. The interaction between the flow of the water-sediment mixture and the sandbed creates different bed configurations which change the resistance to flow and rate of sediment transport. (2) dunes with ripples superposed. river stage.3. (2) antidunes. the river engineer can better analyze alluvial channel flow. the change in bed configuration can cause a three-fold change in resistance to flow and a 10-to-15 fold change in concentration of bed sediment transport. such as the flow depth. Angle of repose of non-cohesive materials (Simons 1955).

resistance to flow is large and sediment transport is small. most movement of the bed material particles is in steps. 1966). consequently. If the antecedent bed configuration is dunes. The velocity of the downstream movement of the ripples or dunes depends on their height and the velocity of the grains moving up their backs. Forms of bed roughness in sand channels (Simons and Richardson 1963.14 . and storage. the changes in depth and slope as the bed form changes. In the lower flow regime. The most common mode of bed material transport is for the individual grains to move up the back of the ripple or dune and avalanche down its face. so the shear stress decreases and the bed form changes to dunes. The bed configuration in the transition zone is erratic. avalanching. Thus. conversely. Resistance to flow is small for flow over a plane bed. Often in the transition from the lower to the upper flow regime. This phenomenon can be explained by the changes in resistance to flow and. depth and slope can be decreased to values more consistent with those of the lower flow regime without changing the bed form. After coming to rest on the downstream face of the ripple or dune. Transition. The dunes cause an increase in resistance to flow which increases the shear stress on the bed and the dunes wash out forming a plane 3. and there is a relatively large separation zone downstream from the crest of each ripple or dune. Lower Flow Regime. depending mainly on antecedent conditions. The water-surface undulations are out of phase with the bed surface. Resistance to flow and sediment transport also have the same variability as the bed configuration in the transition. the depth or slope can be increased to values more consistent with those of the upper flow regime without changing the bed form. if the antecedent bed is plane. It may range from that typical of the lower flow regime to that typical of the upper flow regime. the particles remain there until exposed by the downstream movement of the dunes. they repeat this cycle of moving up the back of the dune. or. The bed form is either ripples or dunes or some combination of the two. the dunes decrease in amplitude and increase in length before the bed becomes plane (washed-out dunes).Lower Flow Regime Upper Flow Regime Figure 3.5.

which covers a wide range of shear values.3. and chutes and pools. and normally the fluid does not separate from the boundary. Different bed-roughness elements may form side-by-side in a cross-section or reach of a natural stream.15 . producing variable roughness. much bed material is briefly suspended. of wave formation and subsidence. The usual bed forms are plane bed or antidunes. giving a multiple roughness. Relation between water surface and bed configuration. antidunes. 3. Upper Flow Regime. then movement stops temporarily and there is some storage of the particles in the bed. or they may form in time sequence. however. Figure 3. when antidunes break. A small separation zone may exist downstream from the crest of an antidune prior to breaking. plane bed with sediment movement. It was the transition zone.6. 1975. The mode of sediment transport is for the individual grains to roll almost continuously downstream in sheets one or two grain diameters thick. For bed materials coarser than 0. dunes form instead of ripples after beginning of motion at small values of stream power. resistance to flow is small and sediment transport is large.2 Bed Configuration The bed configurations (roughness elements) that commonly form in sand bed channels are plane bed without sediment movement. Richardson et al. that Brooks (1958) was investigating when he concluded that a single-valued function does not exist between velocity or sediment transport and the shear stress on the bed.6 mm.6 mm. The relation of bed form to water surface is shown in Figure 3. and of energy dissipation when the antidunes break. In the upper flow regime. dunes.bed. ripples on dunes. The water surface is in phase with bed surface except when an antidune breaks. These bed configurations are listed in their order of occurrence with increasing values of stream power (VγyoS) for bed materials having D50 less than 0.6. Resistance to flow is the result of grain roughness with the grains moving. The different forms of bed-roughness are not mutually exclusive in time and space in a stream. and the cycle continues. 3. ripples.

Change in velocity with stream power for a sand with D50 = 0. velocity. Another example is the change in bed form that occurs with change in the viscosity of the fluid as the temperature or concentration of fine sediment varies over time.19 mm (Simons and Richardson 1966). Variable roughness is related to changes in shear stress. A commonly observed example of the effect of changing shear stress or stream power is the change in bed form that occurs with changes in depth during a runoff event. stream power. and bed configuration is shown in Figure 3. A relation between stream power. or reaction of bed material to a given stream power over time.7.4 m (8-foot) flume at Colorado State University. The relation pertains to one sand size and was determined in the 2. the occurrence of multiple roughness is closely related to the width-depth ratio of the stream. It should be noted that a transition occurs between the dune bed and the plane bed. Figure 3.7. 3. the greater is the probability of a spatial variation in shear stress.Multiple roughness is related to variations in shear stress (γyoS) in a channel cross-section.16 .7). Thus. The greater the width-depth ratio of a stream. In the following paragraphs bed configurations and their associated flow phenomena are described in the order of their occurrence with increasing stream power. either bed configuration may occur for the same value of stream power (Figure 3. stream power or bed material.

the plane bed will change to ripples for sand material smaller than 0.4 Ripples Ripples are small triangle-shaped elements having gentle upstream slopes and steep downstream slopes. the dunes have ripples superposed on their backs. 3.3 m (0. and the flow contains very little suspended bed material. in contrast with ripples. There is a relative roughness effect associated with a ripple bed and the resistance to flow decreases as flow depth increases. 3.012 to 0. 1966) ranged from 0. if any.3. After the beginning of motion. Hence. Prior to the beginning of motion. the relative roughness can remain essentially constant or even increase with increasing depth of flow. With dunes. The ripple shape is independent of sand size and at large values of Manning's n the magnitude of grain roughness is small relative to the form roughness. The bed material discharge concentration is small.6 mm).4 mm. for flat slopes and low velocity. the problem of resistance to flow is one of rigid-boundary hydraulics. Resistance to flow is small for a plane bed without sediment movement and is due solely to the sand grain roughness.4 ft to 2 ft) and height from 0. the problem relates to defining bed configurations and resistance to flow.6 mm.3. disturbance on the water surface. those described by Carey and Keller (1957) in the Mississippi River were 100 to 200 m (300 to 700 ft) long and as much as 12 m (40 ft) high.018 to 0. the bed configuration will be a remnant of the bed configuration formed when sediment was moving.06 m (0. The length of the separation zone downstream of the ripple crest is about ten times the height of the ripple. sand waves called dunes form on the bed. Values of Manning's n range from 0.6 m (0.17 . Resistance to flow is relatively large (with Manning's n ranging from 0. These ripples disappear at larger shear values.014 depending on the size of the bed material. whereas.01 m to 0.5).5). ranging from 10 to 200 ppm. the amplitude of dunes can increase with increasing depth of flow. If the bed material of a stream is not moving.4 m (8-foot) wide flume used by Simons and Richardson (1963. if the bed material is coarser than 0. Dunes are large triangle-shaped elements similar to ripples (Figure 3.3.3 Plane Bed Without Sediment Movement Plane bed without movement has been studied to determine the bed configuration that would form after beginning of motion.06 to 0.12 m to 0.6 to 3 m (2 to 10 ft) in length and from 0. Their lengths range from 0. depending on the scale of the flow system. Dunes that formed in the 2. and to dunes for coarser material. 3.6 m (2 ft) to many tens of meters (hundreds of feet). Ripples cause very little.03 ft to 0.2 ft) (Figure 3.5 Dunes When the shear stress or the stream power is increased for a bed having ripples (or a plane bed without movement.3.2 to 1 ft) in height. Length ranges from 0. particularly if the bed material is coarse sand with D50 > 0. depending on the flow and bed material.5. The bed configurations after the beginning of motion may be those illustrated in Figure 3. At smaller shear-stress values.030). After the beginning of motion. The maximum amplitude to which dunes can develop is approximately the average depth.

01 to 0. 3.3. In natural streams. As the antidunes form and increase in height. the water surface is out of phase with the bed surface (Figure 3. downstream.040. Dunes of fine sand (low fall velocity) are washed out at lower values of stream power than are dunes of coarser sand. The form roughness for flow with dunes is equal to or larger than the sand grain roughness.15 m (0. Resistance to flow caused by dunes is large.5 ft) deep.5 to 2 times the height of the sand waves and the length of the waves.Field observations indicate that dunes can form in any sand channel. This bed configuration is called the transition or washed out dunes. In a flume where the flow depth was about 0.03 to 0. ranged from 1. standing waves. The next bed configuration with increased stream power is plane bed with movement.15 m (0. surface waves 0. from crest to crest. in turn. as with any tranquil flow over an obstruction. Manning's n ranges from 0. With dunes. cause boils to form on the surface of the stream. the dunes elongate and reduce in amplitude. If the sand is coarse and the depth is shallow.010 to 0. however.3. Dunes cause large separation zones in the flow.6 to 1. or antidunes. if the stream power is sufficiently large to cause general transport of the bed material without exceeding a Froude number of unity.5 ft). Manning's n for plane bed sand channels range from 0. With coarse sands larger slopes are required to affect the change from transition to plane bed and the result is larger velocities and larger Froude numbers. transition may not terminate until the Froude number is so large that the subsequent bed form may be antidunes rather than plane bed.7 Antidunes Antidunes form as a series or train of inphase (coupled) symmetrical sand and water waves (Figure 3. 3. 3. they may move upstream. In natural streams.6 Plane Bed With Movement As the stream power of the flow increases further. such as the Rio Grande or the Colorado River.020 to 0.18 .8). and the latter. These zones. Antidunes form as trains of waves that gradually build up from a plane bed and a plane water surface. Their upstream movement led Gilbert (1914) to name them antidunes. much larger antidunes form.5 m (2 to 5 ft) high and 3 to 12 m (10 to 40 ft) long have been observed. The height and length of these waves depend on the scale of the flow system and the characteristics of the fluid and the bed material. the change from transition to plane bed may occur at a much lower Froude number than in flumes. The height of the water waves was 1. In these streams. The waves may grow in height until they become unstable and break like the sea surf or they may gradually subside.6). or remain stationary.3 < Fr < 0. The former have been called breaking antidunes. irrespective of the size of bed material.013. Boundary shear stress in the dune trough is sometimes sufficient to form ripples oriented in a direction opposite to that of the primary flow in the channel. In flume studies with fine sand. Measurements of flow velocities within the separation zone show that velocities in the upstream direction exist that are l/2 to l/3 the average stream velocity. because of their greater depths. the plane-bed condition commonly exists after the transition and persists over a wide range of Froude numbers (0. the height of the sand waves ranged from 0.5).5 to 3 m (5 to 10 ft).

Froude number and regimes of flow and Figure 3. With breaking waves.4 m (8-foot) wide flume at Colorado State University. 3.6 m (1 to 2 ft) per minute.9 conceptualizes the crossover from lower to upper flow regime in natural rivers.2 (Simons and Richardson 1966. Richardson and Simons 1967.018 to 0. These bed configurations are generally called bars and are related to the plan form geometry and the width of the channel (Figure 5. alluvial-channel flow changes to chutes and pools (Figure 3. In larger flumes and in rivers the shift occurs at Froude numbers as low as 0.0 (Fr < 1.012 to 0. Several different types of bars are observed. 3.3. 3. this type of flow and bed configuration was studied using fine sands.0).0) results from studies made in small flumes where the depth is shallow and large velocities are needed in order to shift from a dune bed to a plane bed. In the 2. Dawdy 1961). The flow consisted of a long chute 3 to 9 m (10 to 30 ft) in which the flow was rapid and accelerating followed by a hydraulic jump and a long pool. Nordine 1964. However.19 . The misconception that the lower flow regime shifts to the upper flow regime at a Froude number of 1.10 Bars In natural channels. Manning's n may range from 0. If the antidunes do not break.0 (Fr > 1. some other bed configurations are also found. and Froude Number The change from lower to upper regime flow or the reverse (that is a change from dune bed to a plane bed or plane bed to a dune bed) is not related to the Froude number. standing wave and antidune bed configuration in the upper flow regime only occurs with a Froude number greater than 1. If many antidunes break.5). Richardson 1965. Their shape may vary with changing flow conditions and motion of bed particles but they do not move relative to the bends. resistance to flow is about the same as that for flow over a plane bed. the area of the stream they occupy.3 to 0.3. The elevation of the sandbed varied within wide limits. Bars are bed forms having lengths of the same order as the channel width or greater and heights comparable to the mean depth of the generating flow. Resistance to flow was large with Manning's n of 0.8 Chutes and Pools At very steep slopes.Resistance to flow due to antidunes depends on how often the antidunes form. Figure 3. Bed Configuration. resistance to flow is larger because the breaking waves dissipate a considerable amount of energy.14).035. They are classified as: (1) Point Bars which occur adjacent to the inside banks of channel bends. and the violence and frequency of their breaking.020. The chutes and pools moved upstream at velocities of about 0. and ripples and dunes only occur in the lower flow regime at a Froude number less than 1.0) (Vanoni 1977). 3.0 (Fr = 1.3.8 illustrates the relation between flow depth.9 Regime of Flow.

3.9 Cross-over from lower to upper flow regime based on sand size and Froude Number. Relation between regime of flow and depth for bed material with a median size equal to or less than 0. Figure 3.35 mm.Figure 3.20 .8.

the slope would change and would be dependent on the amount of sediment supplied until a new equilibrium is reached. The variables that describe alluvial channel flow are: V yo Sf ρ ρs µ g Ds G Sp SR Sc fs = = = = = = = = = = = = = velocity depth slope of the energy grade line density of water-sediment mixture density of sediment apparent dynamic viscosity of the water-sediment mixture gravitational acceleration representative fall diameter of the bed material gradation coefficient of the bed material shape factor of the particles shape factor of the reach of the stream shape factor of the cross-section of the stream seepage force in the bed of the stream 3. Portions of the upstream slopes of bars are often covered with ripples or dunes. especially in field studies. with consecutive bars on opposite sides of the channel. These changing effects are discussed along with approximations to simplify the analysis of alluvial channel flow.21 . If a stream is in equilibrium with its environment. In the following sections the variables affecting resistance to flow are discussed. Their lateral extent is significantly less than the channel width. to tell which variables are governing the flow and which variables are the result of this flow. 3. Transverse Bars which also occur in straight channels. and move slowly downstream.4 RESISTANCE TO FLOW IN ALLUVIAL CHANNELS Resistance to flow in alluvial channels is complicated by the large number of variables and by the interdependency of these variables. If for some reason a larger or smaller quantity of sediment is supplied to the stream than the stream is capable of transporting. (3) (4) In longitudinal section. Bars appear as small barren islands during low flows. and Tributary Bars which occur immediately downstream from points of lateral inflow into a channel. the average slope over a period of years has adjusted so that the flow is capable of transporting only the amount of sediment supplied at the upper end of the stream and by the tributaries. with very long gentle upstream slopes and short downstream slopes that are approximately the same as the angle of repose. The effects produced by different variables change under different conditions. Alternate bars move slowly downstream. In such a stream.(2) Alternate Bars which occur in somewhat straighter reaches of channels and tend to be distributed periodically along the reach. It is difficult. They occur both as isolated and as periodic forms along a channel. bars are approximately triangular. The slope of the energy grade line for an alluvial stream illustrates the changing role of a variable. They occupy nearly the full channel width. slope is an independent variable.

Sf.10.CT Cf ω τc = = = = bed material concentration fine-material concentration terminal fall velocity of the particles critical shear stress In general. Figure 3. Most of this presentation is based on laboratory studies which have been supplemented by data from field experience when available. 3. and ripple-bed configuration to dunes. Resistance to flow varies with depth even when the bed configurations do not change. The gravitational acceleration g is also constant in the present context. yo.10. it is possible to experience a large increase in discharge with little or no change in stage. an increase in depth. 3. Also. Nebraska (after Beckman and Furness 1962). a decrease in depth may cause plane bed or antidunes to change to a dune-bed configuration. either with or without sediment movement or ripples. The effect of other variables on the flow in alluvial channels is qualitatively discussed in the following sections. Often there is a gradual change in bed form and a gradual reduction in resistance to flow and this type of change prevents the break in the stage-discharge relation. When the bed configuration is plane bed. there is a decrease in resistance to flow with an increase in depth. and a dune bed to a plane bed or antidunes. and bed material. can change a plane bed (without movement) to ripples. Ds.4. development of dependable stage-discharge relations in alluvial channels is very difficult. Relation of depth to discharge for Elkhorn River near Waterloo.1 Depth With a constant slope. This is a relative roughness effect.22 . For this and related reasons. Nevertheless. analysis of river problems is confined to flow of water over beds consisting of quartz particles with constant ρs. A typical break in a depth-discharge relation caused by a change in bed form from dunes to plane bed or from plane bed to dunes is shown in Figure 3.

The relation between stream power. which in turn determines the fluid velocity and stream power.4. amplitude. 3.2 Slope The slope.06 m/s (0. For example. then decreases as depth is increased further. and activity of the antidunes as depth is increased. depending on the size of bed material and magnitude of the depth.11. When the bed configuration is antidunes. with shallow depths and the ripple-bed configuration. resistance to flow increases with an increase in depth to some maximum value. the effect is uncertain. velocity and bed configuration has been illustrated in Figure 3.4. resistance to flow is affected by a change in slope. Even when bed configurations do not change. Figure 3. an increase in slope increases resistance to flow for bed materials having fall velocities greater than 0.20 ft/s). Additional studies are needed to define the variation of resistance to flow for flow over dune beds. Sf. 3.3 Apparent Viscosity and Density The effect of fine sediment (bentonite) on the apparent kinematic viscosity of the mixture is shown in Figure 3.20 ft/s).7. field and laboratory studies indicate that resistance to flow may increase or decrease with an increase in depth. For those bed materials having fall velocities less than 0. The magnitude of the effect of fine sediment on viscosity is large and depends on the chemical make up of the fine sediment. With the dune-bed configuration.06 m/s (0.23 . Apparent kinematic viscosity of water-bentonite dispersions (Simons and Richardson 1966). resistance to flow increases with an increase in slope. 3.11. This increase or decrease in flow resistance is directly related to changes in length.When the bed configuration is dunes. The slope provides the downstream component of the fluid weight. is an important factor in determining the bed configuration which will exist for a given discharge.

For comparative purposes. ω.12a.12b. 3.24 . Figure 3.12. fine sediment suspended in water increases the mass density of the mixture (ρ) and. γ= γw γs γ s − Cs (γ s − γ w ) (3.4 Size of Bed Material The effects of the physical size of the bed material. consequently.13) where: γw = γs = Cs = Specific weight of the water [9810 N/m3 (62. has a specific weight (γ) of about 10.4. the specific weight (γ). It is clear any change in γ affects the boundary shear stress and the stream power.5 lb per cu ft).In addition to increasing the viscosity. Ds. the effect of temperature on the fall velocity of two sands in clear water is shown in Figure 3. Changes in the fall velocity of a particle caused by changes in the viscosity and the fluid density resulting from the presence of suspended bentonite clay in the water are shown in Figure 3.4 lb per cu ft)] Specific weight of the sediment [about 26. The specific weight of a sediment-water mixture is computed from the relation.4 lb per cu ft)] Concentration by weight (in fraction form) of the suspended sediment A sediment-water mixture. which is a measure of the interaction of the fluid and the 3. Variation of fall velocity of several sand mixtures with percent bentonite and temperature (Simons and Richardson 1966).000 N/m3 (165. on resistance to flow are: (1) its influence on the fall velocity.460 N/m3 (66. where Cs = 10 percent.

Nebraska has bed roughness in the form of dunes in the summer when the water is warm and less viscous but has a nearly plane bed during the cold winter months. and slope. except for the contribution of the grain roughness. two sets of data collected by Harms and Fahnestock (1965) on a stable branch of the Rio Grande at similar discharges show that when the water was cold. Similarly. The physical size of the bed material. That is. Observations of natural streams have shown that the bed configuration and resistance to flow change with changes in fall velocity when the discharge and bed material are constant. the bed roughness was dunes. the depth was relatively shallow.5 Size Gradation The gradation.6 Fall Velocity Fall velocity.particle in the formation of the bed configurations. resistance to flow is about the same for both uniform or graded sand. the resistance to flow. knowledge of the shape factor and density of the particle are also required. Also the transition from upper flow regime to lower flow regime occurs over a narrower range of shear values for the uniform sand. G. but when the water was warm. For a plane bed with motion. 3. the fall velocity determines the bed form that will occur. Flume experiments indicate that uniform sands (sands of practically the same size) have larger resistance to flow (except plane bed) than graded sands for the various bed forms. the resistance to flow was small. 3. The physical size of the bed material determines the friction factor mainly for the plane-bed condition and for antidunes when they are not actively breaking. Use of the fall diameter instead of the sieve diameter is advantageous because the shape factor and density of the particle can be eliminated as variables. The physical size of the bed material for a dune-bed configuration also has an effect on resistance to flow. and the velocity was large. the actual dimensions of the bed form and. which increases with a decrease in the fall velocity of the bed material. the fall velocity of the particle in any fluid at any temperature can be computed. the resistance to flow was large. causes additional dissipation of energy. Ks. although the dissipation of energy by the form roughness is the major factor. The form of the dunes is also related to the fall velocity of the bed material. the depth was large. is the primary variable that determines the interaction between the bed material and the fluid. whereas. For example. and (3) its effect on the turbulent structure and the velocity field of the flow. The flow of fluid over the back of dunes is affected by grain roughness. (2) its effect on grain roughness. the bed of the stream was plane. The breaking of the waves. and the velocity was low. the Loup River near Dunning. as measured by the fall diameter or by sieve diameter. For a given depth. to do the same computation when the sieve diameter is known. ω.4. is a primary factor in determining fall velocity. if only the fall diameter is known. of sizes of the bed material affects bed form and resistance to flow.25 .4. 3.

During the falling stage.8 Seepage Force A seepage force. 3. the stability of the bed material. velocity distribution and resistance to flow. this process increases the stability of the bed and bank material but stores water in the banks. the seepage forces can influence the form of bed roughness and the resistance to flow for a given channel flow. transition.9 Bed Material Concentration The bed material concentration. 3. If there is inflow. As a direct result of changing the effective weight. flow phenomena. Also. In narrow channels dunes and antidunes vary mainly in the downstream direction and resistance to flow is larger than for a wide channel. the forms of bed roughness would be transition. and antidunes. cf. The seepage flow affects the alluvial channel phenomena by altering the velocity field in the vicinity of the bed particles and by changing the effective weight of the bed particles. Sc. bed configuration. consequently. Also.4. in wide channels more than one bed form can occur in the cross-section. plane bed. SR.3. The presence of sediment in the flow causes a small change in the turbulence characteristics. its effect on the viscosity of the water-sediment mixture should be taken into 3. A common condition is outflow from the channel during the rising stage.5 mm will be molded into the following forms as shear stress is increased: dunes. fs. inflow to the channel reduces the effective weight and stability of the bed and bank material and influences the form of bed roughness and the resistance to flow.3 mm).26 . affects the fluid properties by increasing the apparent viscosity and the density of the water-sediment mixture. If there is outflow. Study of these losses in natural channels has long been neglected.4. the effect of the sediment on viscosity µ and the density ρ in any resistance to flow relation is accounted for by using their values for the water-sediment mixture instead of their values for pure water. If significant amounts of sediment is in suspension. CT. and the shape of the cross-section. affect the energy losses resulting from the nonuniformity of the flow in a natural stream caused by the bends and the nonuniformity of the banks. Seepage may have a significant effect on bed configuration and resistance to flow.7 Shape Factor for the Reach and Cross-Section The configuration of the reach. Ds = 0. and antidunes for the same range of flow conditions. However.10 Fine-Material Concentration Fine-material concentration. If this same material was subjected to a seepage force that reduced its effective weight to a value consistent with that of medium sand (median diameter. under shallow flow a bed material with median diameter of 0. the seepage force acts in the direction of gravity and increases the effective weight of the sand and the stability of the bed material. 3. or washload is that part of the total sediment discharge that is not found in appreciable quantities on the bed. and resistance to flow vary with the width of the stream. the seepage force acts to reduce the effective weight of the sand and. For example. the situation is reversed.4. plane bed. occurs whenever there is inflow or outflow through the bed and banks of a channel in permeable alluvium.4.

Texas.27 .13. median fall diameter. Figure 3. Mexico (Culbertson and Dawdy 1964). Observations by the authors on natural sandbed streams with bed material having a median diameter ranging from 0.1 mm to 0.040 at low flow to as small as 0. These observations are substantiated by Dawdy (1961).14.account. This relation gives an indication of the form of bed roughness one can anticipate if the stream power and fall diameter of bed material are known. Nebraska (Beckman and Furness 1962). N. and form roughness is shown.64 mm. (3) Middle Loup River at Dunning. Manning's n changes from values as large as 0. The effect of fine sediment on resistance to flow is a result of its effect on the apparent viscosity and the density of the water-sediment mixture. The method of defining and treating the fine-material load computations is discussed in Chapter 4.S.012 at high flow. the fine sediment is uniformly distributed in the stream cross-section. The lines dividing dunes and transition and dividing transition and upper regime are based on flume data and the following field data: (1) Elkhorn River. Army Corps of Engineers (1968) and Beckman and Furness (1962).4. median fall diameter of bed material. (4) Rio Grande at Cochiti. 3. Relation between stream power. Flume data were utilized to establish the boundaries separating plane bed without sediment movement and ripples.11 Bedform Predictor and Manning's n Values for Sand-Bed Streams In Figure 3. 32 km (20 mi) above El Paso. Colby (1960). and (5) Punjab canal data upper regime flows that have been observed in large irrigation canals that have fine sandbeds. the relation between stream power. An example is given in Figure 3. and bed configuration and Manning's n values.13.4 mm indicate that the bed planes out and resistance to flow decreases whenever high flow occurs.64 mm. 3. Nebraska (Hubbell and Matejka 1959). and plane bed without sediment movement and dunes for D50 coarser than 0. near Bernalillo and at Angostura heading. near Waterloo. and ripples and dunes for bed material with D50 finer than 0. Generally. U. (2) Rio Grande.

4. the height of embankments across the floodplain. and the converse is also true. the maximum dune height may be of the order of 3 m (10 ft). due to the change in bed form that would occur. spur dikes or banks and also increases the required size of riprap. As stated previously large discharges normally experience smaller resistance to flow in a sandbed stream due to the change in bed form. The aggradation and the roughness increases the river stage and thus the height of any control structure or the levees. However.12 How Bedform Changes Affect Highways in the River Environment At high flows. the decrease in stage resulting from the planing out of the bed will decrease the required elevation of the bridge crossing. The resistance to flow is then decreased two to threefold. and half of this would be below the mean elevation of the bed. The corresponding increase in velocity can increase scour around bridge piers. If the depth of flow is 3 m (10 ft). Thus. if the bridge crossing or encroachment causes appreciable backwater. The average height of dunes is approximately 1/2 to 1/3 the average depth of flow and the maximum height of a dune may approach the average depth of flow. most sandbed channel streams shift from a dune bed to a transition or a plane bed configuration. Another effect of bed form on highway crossings is that with dunes on the bed there is a fluctuating pattern of scour on the bed and around the piers.Figure 3.5 m (5 ft) in the local scour would be anticipated when the trough of the dune arrives at the bridge.28 . On the other hand. the dune bed may not plane out at large discharges and a higher resistance to flow results. Change in Manning's n with discharge for Padma River in Bangladesh. With the passage of this dune through a bridge section. abutments. abutments. and the height of any channel control works that may be needed. 3. At low flow the bars can be residual and cause high velocity flow along or at a pier or abutment or any of the other structures in the stream bed. guide banks and spurs. This increase in resistance to flow can decrease the velocity of flow and also decrease the transport capacity of the channel so that aggradation occurs upstream of the crossing. an increase of 1. 3. the height of any dikes.14. the bridge crossing can adversely affect the floodplain. A very important effect of bed forms and bars is the change of flow direction in channels. causing deeper than anticipated scour.

care must be taken in analyzing the crossing in order to foresee possible changes that may occur in the bed form and what these changes may do to the resistance coefficient. In general. The general lack of mobility in coarse material channels also means the bed forms do not change as much or as rapidly as in sandbed channels. The phenomenon of armoring in mobile bed channels occurs by the rearrangement of bed material during movement. coarse-material channels are less active and have slower rates of bank shifting than sandbed channels. The river bed forms (dunes) are less important in the hydraulic behavior of coarse bed channels.29 .4. This classification includes all channels with noncohesive bed materials coarser than 2 mm size. Sampling. If it is anticipated that the bed layer will be disrupted at any given stage. even if the maximum size is limited to cobbles (250 mm). The bed is covered by a one particle thick layer of the coarser material underlain by the finer sizes. In sandbed channels. However. Most of the resistance to flow in coarse material channels comes from the grain roughness and from bars. (2) to assess the bed roughness related to the resistance to flow. The behavior of coarse material channels is somewhat different from sandbed channels.02 to 2 mm. The surface sampling can be easily done on the channel bed by counting particles on a grid. In coarse-material channels. and to the stability of the reach and its structures. The analysis of coarse-material channels is also pertinent to highway engineering. 3. The main distinction between the two channels lies in the spread of their bed material size distribution. and (4) to determine the long and short time response of the channel to specific activities.10 to 250 mm.e.13 Alluvial Processes and Resistance to Flow in Coarse Material Streams The preceding discussion of alluvial channel flow is mainly related to sandbed channels. However. a 100-fold size range. the properties of the surface layer are needed. special effort should be made to obtain an objective sample. The bed and bank forming activity in these channels is therefore limited to much smaller intervals of the annual hydrographs than the sandbed channels. to sediment transport.. channels with noncohesive bed materials of size less than 2 mm. 3. as explained earlier in this chapter. it is necessary to take an adequate sample of both the surface and subsurface material. There is a tendency to select too many large particles. (3) to determine the bed material load for a given flow. The purpose of bed material sampling in coarse-material channels is: (1) to determine the conditions of incipient movement. for example. these channels can armor their beds and behave as rigid boundary channels for all except the highest flows.500-fold size range. The absence of finer sizes from the surface layer is caused by the winnowing away of these sizes by the flow.4 mm (an inch) or so is difficult to obtain and such samples may have to be collected from bars and other exposed areas on channel perimeter. i. The scoop sample with bed material sizes larger than 25. Armoring. that is. For objectives (1) to (3). the tendency for channel armoring is more pronounced in coarse material channels as discussed next. The roughness coefficients in coarse material channels are therefore more consistent during the annual hydrographs than in sandbed channels. the bed may consist of particles from 0. the size range of particles may be 0. which is a 2. As the spread of particle sizes available in the bed of coarse-material channels is large.With highways in the sandbed river environment.

In the size distribution analysis of coarse-bed materials.19 has been proposed by Limerinos (1970) and involves flow depth yo as a parameter.30 . For other distributions. the form roughness is the primary component of channel roughness.15) V. Army Corps of Engineers (1991) Ku = 0. the ripples never form and dunes are rare.14) (3.T. volume.18) Equation 3. With coarse material channels the grain roughness is the main component of the channel roughness. Form roughness can be much greater than the grain roughness when the bed forms are ripples and dunes. then the particles coarser than D84 or D90 need to be analyzed with more care. Comparisons of several equations have been reported by Bray (1982) and Simons and Senturk (1992). For other cases. it is sometimes necessary to obtain particle counts by number.0388 Ku = 0. The main type of bed form roughness in such channels is the pool and riffle configuration. special numerical techniques have to be used to transform the number distributions to weight or size distributions. 1968) shows that the channel roughness can be predicted by various forms of Strickler's equation: /6 n = K u D1 x (3.16) Lane and Carlson (1955) Ku = 0. weight or surface areas directly. 3.0482 Ku = 0.0417 Ku = 0. Care must be taken in the interpretation of frequency distribution of part of a sample obtained by sieving. These sizes also require large samples for their determination. It is difficult to determine the channel roughness for such beds.S.17) U. In sandbed channels.0395 (D50 in meters) (D50 in feet) (3. rather than by sieving or visual accumulation tube analysis for a part of the sample.0256 (D75 in meters) (D75 in feet) (D75 in inches) (3. canals and flumes (Anderson et al.046 Ku = 0. Chow (1959) Ku = 0. analysis of data from many rivers. If the objectives of bed material sampling include bed roughness and channel response. Resistance to Flow.0473 Ku = 0. Only if the size distribution in a sample follows log-normal probability distribution can number counts be transferred to distributions by size. (1970) Ku = 0.0342 (D50 in meters) (D50 in feet) Anderson et al. A coarse-material channel may have bed material that is only partly submerged during most of the flows.038 (D90 in meters) (D90 in feet) (3. In coarse-material channels.

1 1.31 .9 1.n= /6 K u y1 o y 1. is to evaluate the Darcy-Weisbach friction factor. with: 1 æ R = 1. as a function of the hydraulic radius. Table 3.9 ç çD f è 84 ö ÷ ÷ ø 1/ 4 (3.0927 This approach should not be applied to river reaches with active beds or significant sediment transport. feet) Manning's n can then be calculated from the relationship between R and f : n = K u R 1/ 6 f (3.113 Ku = 0. Typical Sediment Size Distribution for Gravel-Bed Stream. for gravel-bed material when size distribution curves cannot be obtained from field data.1 ø 1 æ R = 2.0 log ç çD f è 84 (3.19) Ku = 0.16 + 2 log o D 84 (3. D84.08 3.20) ö ÷ ÷ + 1.3 Standard Deviation 0.113 Ku = 0.3.3 to estimate D90.22) (R in meters) (R in feet) Ku = 0. R. (1978) and Bray (1979) have proposed the guidelines in Table 3. and D65.0927 (yo and D84 in meters) (yo and D84 in feet) An alternative approach. Ratio D90/D50 D84/D50 D65/D50 Mean Value 2. valid when R/D50 > 10.46 0.21) Note: R and D84 in same units (meters.36 0. Charlton et al. f.

(3) the type of problem.32 . Other relations give the shear stress in terms of the velocity of flow. Beginning of motion can be related to either shear stress on the grains or the fluid velocity in the vicinity of the grains. In general. velocity or discharge) and sediment size. For the same mean values. 3. these values are called the critical shear stress or critical velocity.2 Theoretical Considerations When the force of the flowing water (as measured by the shear stress or velocity) is less than some critical value. the following topics are discussed: theory of beginning of motion. It may not be sufficient to determine the average value of the critical shear stress or velocity because both quantities are fluctuating. But when the shear stress or velocity over the bed attains or exceeds its critical value. particle motion begins. tables giving observe values between flow variables and sediment size. slope stability. the alluvial bed can be considered as immobile. equations to determine the relation between flow variables (depth.5 BEGINNING OF MOTION 3. (2) the precision with which the critical value is known or can be determined for the particle size. stable channel design.5.3. waves. Where γ is the unit weight of water. In addition. This difficulty is a consequence of a phenomenon that is random in time and space. the observation of particle movement is difficult in nature. When the shear stress is near its critical value. When the grains are at incipient motion. and seepage into or out of the bed or banks affect the beginning of motion. it is difficult to conclude that particle motion has begun. R is the hydraulic radius and S is the slope of the energy grade line. sediment transport. Section 2. The time history of the movement of a particle involves long rest periods.5. The most dependable data available have resulted from laboratory experiments. they may have larger values that act for a sufficiently long enough time to cause a particle to move. Kramer (1935) and Buffington (1999) proposed four levels of motion of bed material. Shields experimental relationship and its modifications. The choice of shear stress or velocity depends on: (1) which is easier to determine in the field. 3. whereas. Then. the depth. it is possible to observe a few particles moving on the channel bottom.1 Introduction The initiation or ceasing of motion of sediment particles is involved in many geomorphic and hydraulic problems including stream stability and scour at highway bridges. and design of riprap. In this section. In wide channels (width equal to or greater than 10 times the depth) R ≈ y. In fact. In sediment transport analysis most equations use critical shear stress. Equations for determining the shear stress on the bed of a stream are given in Chapter 2. and figures for determining the flow or sediment variables at beginning of motion. The beginning of motion is difficult to define. The average shear stress on the boundary is given by τ0 = γ RS. in riprap design critical velocity is generally used.4. the forces on the particle resulting from the flowing water.5. In stable channel design either critical shear stress or critical velocity is used. the bed material of a channel remains motionless. These problems can only be handled when the threshold of sediment motion is fully understood. erosion.

even though there is a diversity in the experimentally determined coefficients. but also are a function of shape of particle. which would indicate antidunes not ripples. 3. the equations presented in this section are theoretically sound and have proven to give reliable results. though positive evidence of its existence has been presented by Liu (1957) and Senturk (1969). Medium movement: The grains of mean diameter begin to move. no evidence has shown that the mean diameter represents the composition of a mixture correctly. 3. This is true. Gessler (1971). Tison (1953). External forces Fn acting at the points of contact between the grain and its neighboring grains. None. 2. However. Vanoni (1977). The grains moving on one square centimeter of the bed can be counted. its position relative to other particles. have studied the problem of initiation of motion. Weak movement: Only a few particles are in motion on the bed. For coarse sediments such as sands and gravels. It is sufficiently vigorous to change the bed configuration. Also. chemical bonding between particles resists the beginning of motion. The studies involve both theory and experimentation. and the form of bed roughness. 2. Many researchers such as Kramer (1935). The forces that resist the entraining action of the flowing water differ depending upon the properties of bed material. the hydrodynamic forces acting upon a grain are just balanced by the resisting force of the particle. and 3. Buffington (1999) provides an excellent review of Shields research. Simons and Richardson (1966). For cohesive bed material (generally silts and clays). Whether or not a plane bed can exist with weak sediment motion is debated.3 Theory of Beginning of Motion The forces acting on an individual particle on the bed of an alluvial channel are: 1. Shields (1935). if increased even slightly the grain will move. critical or threshold conditions are said to have been reached. White (1940). Data available on critical shear stress are based on arbitrary definitions of critical conditions and most definitions used have relied on subjective visual observations. The complexity of the problem explains the diversity of experimental results. The motion is not local in character but the bed continues to be plane. In reality.33 . the movement is occurring in all parts of the bed at all times.1.5. 4. These forces tend to move or entrain the particles. The engineer facing this dilemma of dealing with beginning of motion in a mixture of sediment sizes should analyze the problem very carefully. there is no truly critical condition for initiation of motion for which motion begins suddenly as the condition is reached. Body force Fg due to the gravitational field. General movement: All the mixture is in motion. Under critical conditions. Liu’s observations may have involved such shallow flow that the Froude number was larger than 1. the resisting forces relate mainly to the weight of the particles. Water flowing over a bed of sediment exerts forces on the grains. When the hydrodynamic forces acting on a grain of sediment have reached a value that.

the fluid forces acting on the grain are divided into three components: (1) The form drag component FD (3. Then the external force Fn is Fn = Fg – FB or Fn = (ρ s − ρ) gK 2 D 3 s (3. Under conditions of no flow. Fluid force Ff (lift and drag) acting on the surface of the grain. the body force is: Fg = gρ s K 2 D 3 s (3.26) FB = gρK 2 D 3 s where: CD K1 K2 Ds v Cs τ = = = = = = = Drag coefficient Coefficient associated with the area of the grain subjected to form drag and shear (the term K 1D 2 s represents the cross-sectional area of the grain) Volumetric coefficient of the grain Diameter of the grain Velocity in the vicinity of the grain Coefficient of shear Average viscous shear stress The external forces Fn depend on the values of the fluid and body forces FB.3. For the individual grain.27) 3. The fluid force varies with the velocity field and with the properties of the fluid.23) where: ρs K2 Ds K 2D 3 s = = = = Density of the grain Volumetric coefficient Grain diameter Volume of the grain For convenience. There is no form or viscous drag.34 .25) FV = C SK 1D 2 sτ (3) The buoyant force component FB (3.24) 2 FD = C D K 1D 2 s ρ ( V / 2) (2) The viscous drag component FV (3. The relative magnitude of these forces determines whether the grain moves or not. the fluid force is Ff = FB.

28 and 3. For turbulent flow. When the flow over the grain is laminar the viscous shear force is predominant and the term DsV*/ν is small. The form drag can be written in terms of the shear velocity. the ratio of the forces tending to move the grain to the forces resisting movement is: 2 ρD 2 τo FD s V* ~ = 3 Fn (ρ s − ρ) gD s ( γ s − γ ) D s (3. consider the ratio of the form drag force FD to the viscous shear force Fv. As both the form drag and viscous shear are proportional to the shear velocity. the local velocity.29) Again. Then. but it is the shear velocity for laminar flow.32) When the flow over the grain is turbulent. the form drag is predominant and the term DsV*/ν is large. According to Equations 3.33) 3.35 .30) With this expression for viscous shear. the external force is equal to the submerged weight of the grain.28) The viscous drag is also related to the shear velocity. Thus. For laminar flow: τ=µ dv dy (3.31: 2 FD ρD 2 s V* ~ Fv µD s V* or FD D s V* ~ Fv υ (3.That is.24 reduces to: 2 FD ~ ρD 2 s V* (3. v is directly proportional to the shear velocity V*. the shear force Fv becomes FV ~ µD S V* (3. Equation 3. the Reynolds’ number of particle DsV*/ν is an indicator of the characteristics of the flow in the vicinity of the grain.31) Now. by replacing V with V* and y with Ds we can write τ ≈ µ V* / D s (3.

34) 3. Gessler (1971) gave Figure 3. Figure 3.4 Shields Diagram Many experiments have been conducted to develop an explicit solution of the Shields relation.15) is widely accepted and τc/(γs .γ) Ds and DsV*/ν for the condition of incipient motion has been determined experimentally by Shields and others. The concept of the Shields Diagram (Figure 3. As the relation is determined experimentally. The earliest one is the graphical presentation given by Shields (1935). In functional form the relation (called the Shields relation) is as follows: æ V* D s τ0 = fç c ç υ (γ s − γ) Ds è ö ÷ ÷ ø (3.Recall that V*2 = τo/ρ.γ) Ds (often referred to as Shields parameter) is widely used to determine the shear stress at beginning of motion.15.16 for the relation given in Equation 3. The relation between τo/(γs .5.36 . the relation considers viscous effects. Shields Diagram: dimensionless critical shear stress. 3. For example.34.

11.15) was divided into three regions by Simons and Senturk (1992) as illustrated in the following." The Shields Diagram (Figure 3. 1 (γ s − γ)Ds (3. Simons and Senturk (1992) state the values of τ/(γs .6ν/V*. Buffington (1999) discusses in great detail Shields' research and states the following: "Nevertheless. and the boundary is considered hydraulically smooth (δ is the thickness of the laminar boundary layer. Because Shields neglected to explain which method he used and did not present sufficient data to recreate his calculations. these two passages from Shield’s dissertation offer two definitions of incipient motion (bed-load extrapolation for uniform sizes versus visual observation for mixed grains.γ) Ds were at least twice as large as the critical value.15) by measuring bed-load transport for various values of τ/(γs .37 .0 In the region Ds < 3δ. Region 1: V* D s / ν < 3. where δ = 11. However.35) then (approximately) 3. when the value of τc ≅ 0.16. Shields estimated the portion of the diagram for V*Ds/ν < 2.γ) Ds is considerably below the critical value. throughout his dissertation he discussed his approach and results as being representative of uniform grains (Shields 1936c pp. Chapter 2). the matter of his experimental procedure remains open to debate. and 16). suggesting that he employed bed-load extrapolation (the method he described for uniform sediment. This indirect procedure was used to avoid the implications of the random orientation of grains and variations in local flow conditions that may result in grain movement even when τ/(γs . Shields’ relation for beginning of motion (after Gessler 1971). He did not perform any experiments in that region. According to Shields. Most authors report that Shields determined this relationship (Figure 3. 14.63 ~ 5.γ) Ds and then extrapolated to the point of vanishing bed load.Figure 3. However.

It must be done with considerable care and with knowledge of channel geometry and hydraulic conditions at the cross section and upstream of the selected cross section. Considering F*.γ)Ds is 0. 3. 1987). Region 3: V* D s / ν > 70 to 500 In this region. identifying initiation of particle sizes by utilizing observed values or by trapping particles in motion over a range of discharges is extremely difficult. 1985.and cobble-bed streams.0 < V* D s / ν < 68. Meyer-Peter and Müller (1948) suggest a value of 0.00 ν Region 2: 3..06.0006 m = 0.10. It is suggested by Simons (Simons and Senturk 1992) that collecting data on initiation of particle motion under field conditions permits selection of a more precise value for the particular channel. dunes form on the bed. The minimum value of F* = τc/(γs .030 to 0.0 mm) bed material range from 0.000 for R*. the boundary is completely rough and F* is independent of Reynolds number R* and is equal to ρV*2 = 0. Also.062 mm to 1. ripples do not form.37) The upper limit of R* in Region 3 is subject to discussion. it can be seen that Ds = 0.6 ν (3. the boundary is in a transitional state and δ/3 < Ds < 6 δ.002 ft.047 instead of 0. For coarser bed material (gravels. However.033 and the corresponding value of R* = V*Ds/ν is about 10.V* D s = 1.039 is a good compromise for all sizes (Fiuzat and Richardson 1983.63 ~ 5.032 ~ 0. cobbles and larger) the experimentally determined Shield's parameter ranges from 0. Some researchers have given values as high as 1.36) The Shields Diagram has a form similar to Darcy-Weisbach's resistance coefficient f versus Reynolds number Re.38 .0 ~ 70. Experimental values of the Shields parameter for sand size (0. Ruff et al. For larger diameter particles.06 (γ s − γ)Ds (3. An average value of 0.020 to 0.6 mm = 0.0 In this region. For this region: ks 1 k s V* = δ 11. it is similar in form to the relation between the drag coefficient CD and the Reynolds number Re for cylindrical bodies and to the relation between V*ks/ν and B = f (V*ks/ν) proposed by Nikuradse (1933).047. If Ds is computed from these values of R* and F*. This is particularly true for gravel.

The variation of the critical shear stress for beginning of motion as a function of sediment size given by various investigators is shown in Figure 3. The spread of the representative curves shows the diversity of experimental and theoretical results. In equation form it is as follows: 3. Comparison of critical shear stress as a function of grain diameter (after Chien 1954). Nevertheless.17 (Chien 1954).39 .5 Equations for Flow and Sediment Variables for Beginning of Motion Particles start to move in steady. 3. Figure 3. A useful relation can be developed between the flow velocity.5. uniform flow when the shear stress applied by the flow equals the resistance to movement of the particles. equations. tables and figures given in this section provide useful tools for determining beginning of motion for sediment particles and the design of riprap.17. and resistance and bed material size by equating the applied shear stress to a resistance to motion shear stress. depth.

43) ρ gn2 V 2 2. m/s. ft Unit weight of water (9. the Shields relation (Vanoni 1975) can be used to determine the relation between the critical shear stress and bed material size for beginning of bed material movement. ft/ft Average velocity.800 N/m3. is as follows: τ0 = γ R S f (3.38) where: τ0 τc = = Average bed shear stress.39) Using y for the hydraulic radius (R) and the Manning equation to determine the slope (Sf) The average shear stress can be expressed as follows: τ0 = ρ g y S f = ρ gn2 V 2 y 1/ 3 (SI) (3. ft/s Diameter of smallest non transportable bed material particle.40 .49) 2 y 1/ 3 (English) (3.22 y 1/ 3 = K s (ρ s − ρ) g D s (English) (3.4 lb/ft3) Manning's roughness coefficient Shield's coefficient 3.τ0 = τc (3.42) At the beginning of sediment movement the applied shear stress is equal to the critical shear stress as given in Equation 3. derived in Chapter 2.36 (τ0 = τc) resulting in the following: ρ gn2 V 2 y 1/ 3 = K s (ρ s − ρ) g D s (SI) (3. N/m2.40) τo = ρ gn2 V 2 (1.44) where: y Sf V Ds γ n Ks = = = = = = = Average depth of flow. 62. lb/ft2 The average bed shear stress applied by the steady uniform flow. m. The relation is as follows: τ c = K s (ρ s − ρ) g D (3. ft Slope of the energy grade line. m. lb/ft2 Critical bed shear stress at incipient motion.41) For noncohesive bed materials. m/m. N/m2.

depth. Critical velocity for beginning of sediment movement for a given depth.4 lb/ft3) Acceleration of gravity (9.44 to give the critical velocity for beginning of motion of bed material of size D for depth y. Critical size for a given velocity. The equation can be solved for the following: 1. and Manning's n results in: Vc = /2 K1 s (S s − 1) n 1/ 2 /2 D1 y 1/ 6 s (SI) (3.48) (SI) (English) (3.94 slugs/ ft3) Density of sediment (quartz 2.49) Ku = Ku = /2 1.041 Knu = 0.647 Kg/m3.49 k 1 (S s − 1)1/ 2 D1 y 1/ 6 s s n (English) (3. Shield's parameter.41 . roughness.43 and 3. This coefficient is called the Shields coefficient. ft/s Shields parameter 3.45) Vc = /2 /2 1. 165. Shields coefficient.43 and 3. 32. 2. Critical Velocity for the Beginning of Bed Material Movement. 1. Clear-water scour depth for a given velocity. 3. roughness. roughness.2 ft/s2) The relationships in Equations 3. Rearranging Equations 3.0336 /3 Vc = K u D1 y 1/ 6 s /2 K1 s (S s − 1) K nu 1/ 2 (SI) (English) (3. density and a coefficient determined experimentally for the beginning of movement of the sediment particles. m/s.44 are the fundamental relations between velocity.Ss ρ ρs g = = = = Specific gravity (2.65 for quartz) Density of water (999 kg/m3.49 K 1 (S s − 1)1/ 2 s K nu where: Vc Ks = = Critical velocity above which bed material of size D and smaller will be transported. resistance to flow (Manning's n). and bed material size and density. Shields coefficient. depth.81 m/s2. Shields coefficient. and bed material density. and bed material size and density. This depth is the contraction scour depth at the end of a long contraction.46) (3.47) /6 n = K nu D1 s Knu = 0.

25 3 (3.42 .25 Critical Size for the Beginning of Bed Material Movement. specific density. Shields parameter and bed roughness. Rearranging Equation 3.78 x 10 −5 K yu = 4.65. size of bed material. This is particularly true if the velocity is converted to discharge using the continuity equation.039. m. depth y. Shield's parameter and Manning's n results in: Dc = Ku V3 y 1/ 2 1 = 0.0336 Ku = 11.94 x 10 −7 (SI) (English) 3.19 3 1 = 0. y= V6 6 Ku D2 s = K yu V 6 D2 s (3. ft Depth of flow.65.041 Ku = 6.48 to give the critical bed material size Ds for beginning of motion for velocity V. m. Knu = 0. for the depth y for no bed material movement for a given velocity.0007 11. Knu = 0.0042 6.19 English Units (ft) Using Ks = 0.50) Ku = (SI) Ku = (English) Depth of Flow For No Bed Material Movement as a function of V and D. Ss = 2.48 can be obtained: Metric (SI) Units (m) Using Ks = 0. Equation 3.039.Ss Ds y n = = = = specific gravity of the bed material Size of bed material.48 can be solved. This equation is useful to determine clear water-scour at a contraction.51) K yu = 1. Ss = 2. ft Manning's roughness coefficient The following values for Ku in Equation 3.

5 gives non-scour velocities for noncohesive and compact cohesive soils suggested by Keown et al.91 1.50 6.52 1.76 0. Original Material excavated for canals Fine sand (colloidal) Sandy loam (noncolloidal) Silt loam (noncolloidal) Alluvial silt (noncolloidal) Ordinary firm loam Volcanic ash Fine gravel Stiff clay Graded loam to Cobbles (noncolloidal) Alluvial silt (colloidal) Graded.9 m (3 ft)] Water transporting noncolloidal silts. Replacing V with Q in Equation 3.50 6.14 0.00 0. From continuity V = Q/(W y) The relation between y.76 0.61 0.02 0.07 1.00 3.75 3.25 2.50 6.22 1.50 3.98 1. no Water transporting sands.00 5.53 0.02 0.035 0.50 2.50 5.025 0.00 6.61 0.98 1.00 3.00 2.50 2.025 3.00 2.52 1. Table 3.02 0.07 1.00 3.76 0.61 0.4.52 n 0.00 5.75 4.68 1.75 3.00 5.68 1.025 0.52) (SI) (English) /7 Ku = K1 yu = 0.00 2.02 0.00 5. Table 3.03 0.210 /7 Ku = K1 yu = 0. after aging of canals [y ≤ 0. The velocity V in Equation 3.50 2. silt to cobbles (colloidal) Coarse gravel (noncolloidal) Cobbles and shingles Shales and hard pans Mean velocity.02 0.46 0.50 3.00 6.75 3.76 1.126 3.025 0.69 0.52 1. for channels at small slope for bed material ranging from fine sand to cobbles.6 Tables for Determining Critical Velocities Table 3.43 .00 2.00 5.00 0.50 3.02 0.52 1.22 1.00 5. Maximum Permissible Velocities Proposed by Fortier and Scobey (1926).52 0.Depth of Flow for No Bed Material Movement as Function of Q and D.5.14 1.50 5.50 2.61 0.61 0. gravels or rock detritus colloidal silt fragments ft/sec M/sec ft/sec m/sec ft/sec m/sec 1.00 5.00 0.02 0.00 5.61 1.14 1.14 1.50 3.4 gives the permissible velocity recommended by Fortier and Scobey (1926).07 1.76 0.83 1.52 1. (1977) for bed material ranging from loess soils to boulders.91 1. Q and D for the condition of no bed material transport is very useful in determining clear-water contraction scour as explained in Chapter 7.91 1.00 2.03 0.51 can be replaced with the discharge Q using the continuity equation (Q = W y V).51 to determine the depth (y) when there is no bed material movement of size D.52 1.83 2. gives the following equation: y = Ku Q6 / 7 D2 / 7 W 6 / 7 (3.75 2. Clear water.46 0.50 1.00 4.83 1.

0-1.0 2.0131 0.44 .2 4.0-2.4 4.000820 0.2 8.840 0.00164-0.1 3.4 8.2 .7 2.00328-0.2 5.1 1.9 3.Table 3.7 3. causing the particles to aggregate and not act as individual particles.9 4.2 1.210 0.105-0.8 1.8 4.420 0.9 20.6 ft 9.18.9 3. 1977).7 13.9 4.5 5.98 3.8 7.50 Medium sand 0.5 2.2 4.3 2.105 0.0 Fine gravel 8.3 19.25-0.4 10.0525 0.3 ft 3.8 3. 3.1 4.000410 Approximate Nonscour Velocities (feet per second) Mean Depth 1.3 3.8 ft 15.3 16.420-0.4 11.0 Very coarse sand 2.8 1.6 16.1 2.6 3.6 4.3 ft 6.6 4.2 6.0131-0. Kind of Soil (mm) For Noncohesive Soils Boulders Large cobbles Small cobbles Very coarse gravel Coarse gravel >256 256-128 128-64 64-32 32-16 Grain Dimensions (ft) >0.210-0.0 Very fine gravel 4.18 shows the relationship between critical shear stress and mean diameter as determined and/or recommended by different investigators for different soil types.8 2.9 6.1 11.125 For Compact Cohesive Soils Sandy loam (heavy) Sandy loam (light) Loess soils in the Conditions of finished Settlement 3.5 1.25 Fine sand 0.50-0.0-4.840-0.2 7.1 2.6 2. Figure 3.00656-0. The difference between investigators is possibly due to the effects of cohesion.0-0.3 Medium gravel 16-8.0525-0.00164 0.3 2.1 3.1 1.0 Coarse sand 1. when present.000820-0.3 3.0262-0.7 Figures for Determining Critical Shear Stress or Velocity Figure 3.0 15.5.00656 0. Nonscour Velocities for Soils (Modified from a report by Keown et al.5 1. Critical shear stress as a function of grain diameter (after Lane 1953).8 1.5.6 3.4 2.0262 0.3 2.00328 0.0 1.7 3.

45 .19. To solve the equations Manning's n.65. Ss.8 Summary Equations 3. Equation 3. and Manning's n to fit their specific data.039. In these basic equations. specific gravity. uniform flow.0336 D s for English units. 3. These equations can be used to determine the critical size. (2) specific gravity Ss = 2 . They are for steady. is a function of V6 and Q6/7. and Shield’s parameter are substituted to obtain equations for the dependent variables. Figure 3. DC. respectively. and critical depth for the beginning of bed material movement based on the Manning equation.45 and 3. Note that the critical sediment size. Critical velocity as a function of stone size. and (3) Manning's n = 0.041 /6 1/ 6 D1 The derivations are given in s for metric units and n = 0. and Shield’s parameter must be determined as well as the other variables. reasonable values of Manning's n.Figure 3. specific gravity of the bed material and Shield’s parameter in metric and English units.52 determines the depth of flow as a function of discharge in metric and English units for these same conditions.19 gives the relationship between maximum allowable velocity (critical velocity) against stone or minimum stone size that can sustain hydraulic forces without motion. is a function of V3 and the critical depth.5. sufficient detail that engineers can substitute other values for Ks. specific gravity. These values are (1) Shield's parameter Ks = 0.46 determine critical velocity. 3. In addition. y.

the theory. 3. Sediment transport is the time rate of a quantity of sediment (by weight) moving past a cross-section of the stream. and wash load.20. bed material load. Definition of sediment discharge (load) components (Lagasse et al. 2001).6.6 SEDIMENT DISCHARGE MEASUREMENT 3. In Chapter 4. SUSPENDED BED MATERIAL LOAD Composed of particles typically found in the bed that are transported in suspension. BED LOAD Composed of particle sizes found in the bed that move by surface creep. (Section 4. Because it is measured by weight (often as tons). BED MATERIAL LOAD TOTAL SEDIMENT LOAD Figure 3.46 . 3. The terms are defined in Chapter 4. discharge is often referred to as sediment load.20 from HEC-20 (Lagasse et al. The many terms describing sediment transport result from using or misusing and mixing terms that describe (1) the source of the sediment (bed material load and wash load).3. bed load.2 Terminology There are many terms that have developed over time to describe the many aspects of the transport of sediment by water. equations. Other terms associated with sediment transport are suspended sediment discharge. Washload moves in suspension and is provided by bank and watershed erosion. sliding. 2001). Velocity Profile Sediment Concentration Profiles WASH LOAD Composed of particles finer than those found in appreciable quantities in the bed. (2) mode of particle movement (suspended bed material load and bed load) and (3) measurement of sediment discharge (measured sediment discharge and unmeasured sediment discharge). bed material discharge.1 Introduction In this section the basic terms and methods of measuring sediment discharge (sediment load) are described. saltation or rolling within the boundary layer.6. and methods of computing bed material transport are described. It is a discharge.2) and illustrated in Figure 3.

very coarse sediment. The measurement of suspended-sediment discharge is described in detail in Techniques of Water-Resources Investigations of the United States Geological Survey (Guy 1970. they vary across the stream. Develop a time suspended-sediment concentration graph similar to the stage hydrograph at a gaging station (suspended sediment hydrograph). Vanoni 1977. mg/l Coefficient to convert to metric or English tons per day The determination of water discharge was described in Chapter 2.3 ft) above the bed (Figure 3. Determine the daily suspended-sediment discharge in English or metric tons per day. Richardson 1994) developed samplers that take a velocity averaged suspended-sediment concentration in the vertical to 0. Standard stream gaging procedure described in Chapter 2. Measure the velocity weighted mean suspended-sediment concentration of the flow.6. 4. 1943. Whereas. The Federal Government through the Interagency Subcommittee on Sedimentation (U. 3. 1977.c.3 Suspended Sediment Discharge Measurement The measurement of the suspended sediment discharge of a stream requires the time dependent measurement of the water discharge (discharge hydrograph) and velocity weighted measurement of the concentration of sediment particles moving past the crosssection. Time dependent measurement of the water discharge (discharge hydrograph). In addition to the velocity and concentration varying with depth.b.09 m (0. The essence of the procedure is as follows. 2.3.22). Very fine sediment or coarser sediment in a very turbulent stream may not decrease in concentration with depth (uniformly distributed). Guy and Norman 1970.47 .086 (SI) 0. 1948.21 the velocity decreases with depth and the sediment concentration increases with depth.0027 (English) Qs Qw C ku = = = = Suspended sediment discharge. using Equation 3. In equation form: Qs = K u Q w C (3. 1. metric or English tons per day Water discharge. and 1952. 3. m3/s or cfs Velocity-weighted mean concentration (by weight) of sediment. and Porterfield 1977). 3. Interagency Subcommittee 1940a.6.S.53.4 Velocity Weighted Mean Suspended-Sediment Concentration As illustrated in Figure 3. or finer sediment in placid flow may have a very large increase in concentration with depth.53) Ku = Ku = where: 0. These samplers are used to take samples across the stream or to obtain a coefficient to be applied to a sample at a single vertical to obtain the mean discharge weighted concentration in the cross-section. 1941a.b.

Figure 3. 3. Interagency 1952). Schematic sediment and velocity profiles. Suspended sediment sampler-D49 (Guy and Norman 1970.21. U.S.22a.48 . Figure 3.

However.53. Book 3. Using this concentration and the average daily water discharge in Equation 3.5 Suspended Sediment Hydrograph The measured suspended-sediment concentration is used to plot a concentration hydrograph (Figure 3.6. 3.23. 3. Chapters C3 (Porterfield 1977) Computation of Fluvial-Sediment Discharge describes methods to construct a suspended-sediment hydrograph. Suspended sediment sampler-DH48 (Guy and Norman 1973. The suspendedsediment discharge is determined for each time increment using Equation 3. the day will have to be subdivided into short time increments.S. Engineering judgment. the stage hydrograph and knowledge of the stream is used in drawing the suspended-sediment hydrograph. the suspended-sediment discharge for each time increment for the day are converted to a common time base. U.6.6 Determination of Daily Suspended-Sediment Discharge Using the concentration hydrograph an average daily suspended-sediment concentration is determined. If the concentration and water discharge are changing rapidly during the day.22b.49 . The Techniques of Water-Resources Investigations of the United States Geological Survey. Interagency 1952).Figure 3.53 the average daily suspended-sediment discharge is determined. added up and divided by the number of common time increments to obtain the average suspended-sediment discharge for the day. as illustrated in Figure 3. 3. Then.23).

Gage height and suspended sediment concentration hydrograph. 3.50 . Texas.23. Colorado River near San Saba.Figure 3. May 1-6. 1952 (Porterfield 1977).

1956. A suspended-sediment sampling program is established and the unmeasured sediment discharge determined as in 1 above and added to the measured suspended-sediment discharge. These are: 1. and concentration of wash load. Where there is a suspended-sediment sampling program. 3.7 Total Sediment Discharge The suspended-sediment discharge determination described above is the measured sediment discharge. This unmeasured sediment discharge can be as low as 10 percent of the total sediment discharge (or total load) to as high as 50 percent or more of the total sediment discharge. There is the unmeasured sediment discharge composed of the sediment moving in contact with the bed (contact sediment discharge or load) and the suspended sediment discharge that the sampler doesn’t sample. Here.3. The total sediment discharge is determined by the modified Einstein method described in Chapter 4 for a range of discharges and a sediment rating curve developed. In some stream reaches the turbulence is so large that the total sediment discharge is in suspension and each size fraction of the total sediment discharge is uniformly distributed in the vertical. By mode of transport: The suspended-sediment discharge and the bed (contact) sediment discharge (Figure 3.51 . the suspended-sediment discharge is the total sediment discharge. Burkham and Dawdy (1980) and others have further modified Colby's procedure. (1955. Computer programs are used to determine the total bed material discharge. The percent of the total sediment discharge that is measured depends on the turbulence of the stream. The wash load discharge may be ignored or estimated from periodic suspended sediment measurements. 2.20). By source: The bed material discharge and the wash load discharge (Figure 3. 4. Many of these streams and in some cases streams where the turbulence was artificially created were used by Colby et al. Streams with very large concentrations of wash load or very large turbulence and fine bed material will have very low percent of unmeasured sediment discharge. bed material size. 1962) to develop the modified Einstein procedure to determine the total sediment discharge of a stream as described in Chapter 4. Some of these are: 1. By measurement: The measured sediment discharge and the unmeasured sediment discharge. the total sediment discharge has three classifications. 2. To determine the total sediment discharge various methods are used. As discussed in Chapter 4.6. The curve and a record of daily water discharge are used to determine daily and yearly total sediment discharge.20). 3. 3. the unmeasured sediment discharge is determined using methods described in Chapter 4 and added to the measured suspended-sediment discharge.

0 1.6.83mm ) 100 Dm = 100 = 0. Dm = Dm = åi=1 pi Dsi 100 n åi=1 pi Dsi 100 n = (0. D84.3 ) (2..4.8 ) (0.. the D16.2.34 mm . which lies in the sand size.7.250 14. Sand Bed Material Size Distribution.002) (.250 .2 From the grain size analysis calculate the following statistics: the geometric mean of each size range. Size Range Percent of Total Weight (mm) in Size Range .7 SOLVED PROBLEMS FOR ALLUVIAL CHANNEL FLOW (SI) 3..2.088mm ) + . Likewise for the remainder of the size ranges.002 .0625 0. D50. 34.2 .2 3.7.44 ) (0.(0.4...00 .125 ...011 mm.4 . The effective diameter (Dm) of the sample distribution is calculated with the use of Equation 3.1.12 in Section 3.00 0.8 .6.011mm ) + (4.00 5.9 . D90 sizes.00 0.3. the results are summarized in Table 3.00 .0625 . and the fall velocity of each size range.500 74.0625)]1/2 = 0. Table 3.52 . effective diameter Dm. the gradation coefficient.125 4.1 PROBLEM 1 Sediment Properties and Fall Velocities Sand Bed Channel The bed material size distribution of a sand bed channel is shown in Table 3. For the first size range: Di = [(0.5 2. The geometric mean is calculated as the square root of the product of the end points of a given size range.500 .

500 .4 0. DI (mm) 0.011 0.707 1.. For the geometric mean size of the first size range Di = 0. Sand Size Bed Material Properties..83 TOTAL Percent of Bed Material in this Size (pI) 0.1 at 60o F.0 0.00 Geometric Mean Size.24.250 .00 .8 5. The sand size distribution is plotted on log-probability paper in Figure 3.41 2.39 2.0 pIDI 0.0884 m/s etc.002 ..7.1 100 0.335 The fall velocity is determined with the use of Figure 3.00 2.500 .2 19. ω = .6 0.0457 0.00 1.9 5.0001 0.1.Table 3. 3.24 mm 0.354 0.7.011mm.51 26.0884 0.125 .5 0. ω = 0.51 3.3 0.198 0.2 Fall Velocity (m/sec) 0. similarly Di = 0.177 0.8 100.088 0.54 0.4 94.2 1 0.2.4.00 .24 from which the following values can be obtained: D50 D16 D90 D84 = = = = 0.53 .0001 m/s.71 0. Bed material size distribution curves. The results are shown in Table 3.01 0.3 99.707 mm.8 4.1 90 80 70 60 50 40 30 20 10 0 Percent Finer Figure 3..46 mm 0.250 .2 74.125 .3 99.2 100 Percent Finer 0.33 mm 0. Size Range (mm) .4 14.0625 .0183 0.56 34.0061 0.44 mm 10 8 6 5 4 3 2 10 Diameter. mm 1 0.8 0.0625 .5 0.

00 58. For the largest geometric mean size: Di = 5.9.00 1.7 1. D50.00 6. D90 sizes.2.8.8 4.12: Dm = åi=1 pi Dsi 100 431. D84.6 .Finally. The geometric mean size is calculated as the square root of the product of the end points of a given size range.8): Table 3. ω = 0.66 mm.500 . the Dm.24mm 0..1 at 60o F.177mm ) + .3 2.(58.66mm ) 100 Dm = 100 = 4.44mm ù + + ê ú= ê ú ë D16 D50 û 2 ë 0. the effective diameter Dm.24) from which the following values were obtained: 3.35 Gravel Bed Channel A gravel bed stream channel was sampled and the grain size analysis yielded the following results (Table 3. the fall velocity of each size range.5 ) (5. the gradation coefficient G is calculated using Equation 3.250 0.00 .54 .66 mm. The gravel bed material size distribution was plotted on log-probability paper (Figure 3. for each geometric mean size. G= 1 2 é D50 D84 ù 1 é 0.32 mm The fall velocity is calculated with the use of Figure 3 .518 m/s. Size Range Percent of Total Weight (mm) in Size Range . Gravel Bed Material Size Distribution.8.1) (0.9 n = (0.00 . The effective diameter of the sample distribution is calculated with Equation 3. with the results shown in Table 3.1.9.1 .500 1.33mm 0.. For the largest size range: Di = [(4) (8)]1/2 = 5.00 31. D16. the results are shown in Table 3.250 .4. Likewise the rest of the size ranges. and the gradation coefficient... Likewise the rest of the size ranges are calculated..00 .9.125 .5 From this size distribution calculate the following statistics: the geometric mean of each size range.33mm û G = 1.

5 mm û G = 1. .088. .192.018 0. .067.131. .067.7 mm ù + + ê ú= ê ú ë D16 D50 û 2 ë 2.7 41.177.5 mm D84 = 6.262. .195.070.55 .232.110. .5 Percent Finer 0.128. .00 2.195.110.226.4 mm D50 = 4.091 0. . .119. .192.9: G= 1 2 é D50 D84 ù 1 é 4. .20 8. .226.159.7 mm D90 = 7.180. .566 1. .082. . . .122.104.043.073. Gravel Bed Material Properties.155. .125 . and G of the bed material. . .244 0.256. .040.110. . .223. .143. .0. .0.098.098.021.015. . .9 Pebble Count The b axis (Intermediate length) of 100 particles were randomly picked up and measured on a line in and along the flow of the Cache la Poudre River.259. .4. .192. . . .244 .159.104.0 pIDI 0.012.0. . .7 3.226.152 0. . . .155.265.436 0. . .177. .034. .00 4.152 .518 TOTAL 100 431.1 1.9.00 .088 0.046.049. .1 mm The gradation coefficient is calculated by Equation 3.241. .192 Determine the D50.226.41 2.162.195. . .305 0. .113.082. .110.0.131. 1 6 14 24 27 20 8 100 Percent Finer 100 94 80 56 29 9 1 3. . .00 .018.207 .128. .707 1.018 0.8 58.177.250 . .305 . Size Range (mm) .0 331.6 1.030 TOTAL Class No.354 0. . First decide on discrete size groups and then determine the number of particles in each group. .174. . .5 mm 6. .137.216. .210.171. .7 6. . The b axis lengths in meters are.256. .192.500 . . . .037. Di (mm) 0. .198 0. .0.082. .3 31.500 . . . .040.201.128. .00 1.D16 = 2. .226. . .028. .046.012 .146.016. . . .198 .104.4 mm 4.00 .195.250 .226.177 0.091 .134. .171.146.5 100.098.436. .88 90. D84.0.146. . .2.049. .0. . .015.235.335 0.140.00 Geometric Mean Size. .8. .1 1.192. .046 0.189.280. The sampling was done by walking in a straight line and picking up and measuring the particle at the big toe at each step.4 9. .055.113. D16.0. D90. . . . 137.113. Size Range (m) 0.83 5. .66 Percent of Bed Material in this Size (pI) 0.113.1.0.046. .030 .082.1 Fall Velocity (m/s) 0. 0.198 0. . .67 Table 3. .

From Figure 3.28 m 0.26 m ù + + ê ú= ê ú = 1.000 mm = 1 m = 3.26 m 0.7. D84. and G of the bed material (Figure 3.2 PROBLEM 2 Angle of Repose Determine the angle of repose for well rounded riprap with a D30 of 1 ft = 305 mm.23 m (190 mm) (120 mm) (280 mm) (260 mm) (230 mm) G= é D50 D84 ù 1 é 0.48 ë D16 D50 û 2 ë 0.19 m û 1000 800 600 500 400 300 200 Diamter (mm) 100 80 60 50 40 30 20 10 100 90 80 70 60 50 Percent Finer 40 30 20 10 0 Figure 3. D16. D50 D16 D90 D84 D75 = = = = = 1 2 0.28 ft). D90. 3.19 m 0.Plot on log-probability paper to determine D50.12 m 0. Size distribution curve for pebble count (1.23 m 0.56 .25.19 m 0. 3.25).4 the angle of repose is 40o.

35 mm.1 mm D84 = 4.7. An estimate of the bed form and n-value of the channel is desired.06) (9800) (24. an average velocity of 1.00 m. and a bed slope of 0.57 .013 (b) Antidune Flow A sandbed channel is observed to have an undulating water surface.3 PROBLEM 3 Resistance to Flow in Alluvial Channels Manning's n in Sand Bed Streams (a) Plane bed Considering only the bed of the stream determine the following: What is the range of Manning's n values for a plane bed sand channel stream? Manning's n ranges from 0.281 = 3.0482 (0.07/32. Determine the stream power.06) (0.07 m3/s.00 x 1.003) = 22.225 lb lb N x x = 152 sec − ft 3. Assuming the sieve diameter equals the fall diameter.126 N/sec-m. Using the Strickler equation as an approximation: n = 0.0482 D1/6 with D in m n = 0. The stream bed has a D50 of 0.013. V = 1.3. a discharge of 24. Based upon the bedform the n-value is estimated to be 0.13 for this stream power indicates that upper flow regime is expected.13 is in English units. D16 = 1.48 ft/sec Stream Power = 22.010 to 0.00032 m)1/6 = 0.06 m/s.9 mm D50 = 3.013.0 mm D90 = 4.32 mm. Stream power equals Vγyo So = (1. a channel width of 32.8 mm D75 = 4. Manning's n in Gravel Bed Streams Use the following gravel bed material size analysis to estimate the Manning's n-value of the stream.4 mm 3. The bed configuration should be antidunes with standing waves.06 x 3. Therefore convert velocity and stream power to English units. and the bed slope equals the friction slope. Figure 3.281ft N sec − N Figure 3.003. What is the Manning's n for a plane bed stream with a D50 of 0.126 1m 0.

113 (1.15) (3.19 m (190 mm) 0.17) (3.113 R1/ 6 f 1/ 2 = 0.90 = 0.0049 ) .0417 D.019 (3.0 log ç æ R ö æ 1.018 is selected.18 ø è è D 84 ø (3.12 m (120 mm) 0.016 = 0.1 = 2.0 log ç 0 ÷ çD ÷ è 84 ø = 0.16 +2.167 = 0. 9 ç ÷ ç è D 84 ø 0.20) f1/2 = 0.21) f1/2 = 0.52) 0.019 (Assume yo = 5.19 (3.0 log ç ÷ è 0.52 ö ÷ ç ÷ + 1.0031) 167 n = 0.52 ö 1.50 = 0.0473 D.113 y0 1/ 6 æ y ö 1.1 = 6.23 m (230 mm) 3.0 log ç 0.0044 ø = 8.018 = 0.015 (3.28 m (280 mm) 0.004 ) 167 n = 0.162 n = 0.25 (3.52 ö = 1.122) = 0.0417 (.046 D. Alternatively.16 + 2.167 = 0.113 (1.0044 ø 0.0031) .0482 (. Manning's n in Cobble Bed Streams What are the Manning's n values for a cobble bed stream with the following size distribution: D50 D16 D90 D84 D75 = = = = = 0.167 = 0.15 through 3.22 167 n = 0.162) = 0.18) .25 æ 1.58 .16) (3.52 m and R = yo) n= 0. the maximum and minimum n value could be used and a decision made on the basis of the velocity and discharge.52)1/ 6 (0.50 = 0.75 = 0.113 R1/ 6 f 1/ 2 = 0.22) Based on the range of n-values indicated above a n-value of 0.0473 (0.25 m (250 mm) 0.113 (1.0044 ÷ + 1.167 .0 ft = 1.122 n = 0.19) 1 f 1/ 2 æ R ö ÷ = 1.019 æ 1.046 (0.167 167 n = 0.0482 D. 9 ç ÷ è 0.22) 1 f 1/ 2 = 2.020 (3.52)1/ 6 (0.Using Equations 3.

3.28 m ) 1/ 6 1/ 6 = 0.31 x 10-6 = 159 This point plots above the incipient motion line in Shields (Figure 3.59 .0417 D1/ 6 with D 50 in meters n = 0.00031)/1.n = 0.00038 = 4.91 V* Ds/ν = (4.γ) Ds = 4.4 PROBLEM 4 Beginning of Motion Bed Material Movement Using Shields Figure To establish if the bed material of a channel is in motion the Shields' relationship can be used. R S D50 D16 D90 D84 = = = = = = 1.31mm 0.0031) = 0.543 / (25970-9800) 0.0482 D1 50 with D in meters n = 0.543/1000)1/2 (0.7.00031 = 0.24mm 0.0417 (0. is 9800 N/m3 x 1.091 V* Ds/ ν= (4.543/1000)1/2 (0.46mm 0.23 m ) /6 n = 0.037 3.046 (0.42mm The shear stress on the bed. Ds = 3.037 = 0.046 D1 90 with D 90 in meters n = 0.17) (3.0482 (0.032 (3.16) (3.543 N/m2.037 /6 n = 0.31 x 10-6 = 15.15) (3. (b) For the gravel sized material with the same values of τo = 4.543 /(25970-9800) (0.0473 (0.00038 m/m 0.543 N/m2 τo/(γs .19 m ) 1/ 6 = 0.18) 1/ 6 = 0.γ) Ds = 4.15) and indicates the bed of the channel is in motion.15) and indicates this channel bed is in motion.1 mm we have: τo/(γs .22 m 0. Recall that τo = γRS and V* = τ o / ρ (a) For the following conditions determine if the bed material in a sand bed channel is in motion.22 m x 0. at a single vertical in the cross-section.0473 D1 75 with D 75 in meters n = 0.95 This point plots above the incipient motion line in Shields (Figure 3.19 m) /6 n = 0.0031)/1.

2.19 x 3.66 1/ 6 x 0.52 m / s (b) Gravel Size Bed Material Given depth y is 3.0042 V 3 y 1/ 2 = (0. 3.19 x 3.00399 m (3. Determine the critical bed material size D50.19 x 3.50 mm ) V = 1.0005 m (0.0042) 0.66 1/ 2 = 0.12 m / s (c) Cobble size Bed Material Given depth y is 3.0042) 2.19 y 1/ 6 D1/ 3 = 6.66 m and velocity V is 0.19 y 1/ 6 D1/ 3 = 6.44 3 3. 1.66 1/ 6 x 0.22 m/s.0042 V 3 y 1/ 2 = (0.00311/ 3 = 1. V = 0.128 1/ 3 = 3.Critical Velocity for Beginning of Bed Material Movement (a) Sand Size Bed Material Given depth y is 3.9 mm ) V = 3.0042 V 3 y 1/ 2 = (0.0319 m (31.19 y 1/ 6 D1/ 3 = 6.66 and bed material size D50 is 128 mm.66 1/ 2 = 0. respectively.22 3 3.66 m/s Dc = 0.31 mm.88 m/s.000311/ 3 = 0. what is the critical velocity Vc? Vc = 6.1 mm.44 m/s Dc = 0.99 mm ) V = 2. and 4.66 m/s. what is the critical velocity Vc? Vc = 6.66 m and bed material size D50 Is 3.66 1/ 2 = 0.613 3.108 m (108 mm ) 3.88 m / s Critical Size for Beginning of Bed Material Movement Given depth y is 3. what is the critical velocity Vc? Vc = 6.60 .0042) 1.61 m/s Dc = 0.66 m and bed material size D50 is 0.66 3 3.0042 ) 3.22 m/s Dc = 0.61 m/s.661/ 6 x 0.44 m/s.0042 V 3 y 1/ 2 = (0.66 1/ 2 = 0.

00004 m There would not be any bed material movement at this velocity.78 x 10 −5 ) 2.0042 V 3 y 1/ 2 = (0.66 1/ 2 = 0.61.88 m/s Dc = 0. V = 2. determine the critical depth y when bed material movement would stop.44 m/s y= 1.88 m/s and bed material size D50 of 152 mm.152 2 = 10.44 6 0.616 0.86 m (b) Given velocity V of 0.152 2 = 0. 3.7.78 x 10 −5 V 6 D2 = (1.16 m V = 4.78 x 10 −5 V 6 D2 = (1.8. Why not? The coarser material would armor the bed. Therefore.78 x 10 −5 ) 0.61 m/s and bed material size D50 of 0. 3. sediment sizes are given in mm for the English system. V = 0. y = 1. determine the critical depth y for no bed material movement.88 m/s y= 1.78 x 10 −5 V 6 D2 = (1.266 m (255 mm ) Critical Depth When the Bed Material Movement Would Stop (a) Given velocity V of 0.1 PROBLEM 1 Sediment Properties and Fall Velocities Generally.000305 2 = 9.88 3 3.61 m/s y= 1.152 2 = 0.0042) 4.V = 4.78 x 10 −5 ) 0.78 x 10 −5 ) 4.88 6 0.8 SOLVED PROBLEMS FOR ALLUVIAL CHANNEL FLOW (ENGLISH) 3.44.616 0. the solution to this problem is identical to the SI solution in Section 3. and 4. 2.4 m Would this depth occur? No.305 mm.1.78 x 10 −5 V 6 D2 = (1.61 .

an average velocity of 3. The stream bed has a D50 of 0.003) = 1.9 mm D50 = 3. a channel width of 105 ft. An estimate of the bed form and n-value of the channel is desired.48 ft/sec.32 mm.8 mm D75 = 4.3.15 through 3.0010 ft)1/6 = 0.8.1 mm D84 = 4.53 lb/sec-ft. n = 0.62 . Manning's n in Gravel Bed Streams Use the following gravel bed material size analysis to estimate the Manning's n value of the stream.010 to 0. The bed configuration should be antidunes with standing waves. a discharge of 850 cfs.22 3.4 mm Using Equations 3.48) (0. 3. From Figure 3.003.2 PROBLEM 2 Angle of Repose Determine the angle of repose for well rounded riprap with a D30 of 1 ft = 305 mm. Figure 3. Based upon the bedform the n-value is estimated to be 0.13 for this stream power indicates that upper flow regime is expected. Determine the stream power. D16 = 1. and a bed slope of 0. and the bed slope equals the friction slope.8.48) (62.4) (850/105 x 3.013 (b) Antidune Flow A sandbed channel is observed to have an undulating water surface. What is the Manning n for a plane bed stream with a D50 of 0.0395 (0.4 the angle of repose is 40o. Stream power equals Vγyo So = (3. Assuming the sieve diameter equals the fall diameter.013.3 PROBLEM 3 Resistance to Flow in Alluvial Channels Manning's n in Sand Bed Streams (a) Plane bed Considering only the bed of the stream determine the following: What is the range of Manning's n values for a plane bed sand channel stream? Manning's n ranges from 0.0 mm D90 = 4.0395 D1/6 with D in ft.013. Using the Strickler equation as an approximation: n = 0.35 mm.

18 è ø ø f1/2 = 0.038 D .019 167 n = 0.17) (3. 9 ç çD è 84 ö ÷ ÷ ø = 0.016 167 n = 0.82 ft 0.50 = (0.0)1/ 6 (0. Manning's n in Cobble Bed Streams What are the Manning's n values for a cobble bed stream with the following size distribution: D50 D16 D90 D84 D75 = = = = = 0.122 n = 0.62) /6 1/ 6 n = 0.167 n = 0.0161).0388 ) (0.75 ft (190 mm) (120 mm) (280 mm) (250 mm) (230 mm) (3.75 = (0.0927 (5.019 (3.162) = 0.0) 0.167 = 0.0102 ).63 . Alternatively.50 = (0.0 log ç çD è 84 ö æ 5 .0395 ) (0. n = 0.167 = 0.015 1 f 1/ 2 (3. 0 ö 1.16 +2.19) 1 f 1/ 2 ö æ 5.0102).9 ç 0.0927 (5.0 ft and R = yo) n= /6 0.16) (3.395 D.0 log ç 0.0927 R1/ 6 f 1/ 2 = 0.0)1/ 6 (0.0144 ÷ + 1.15) (3.0144 ø 0.39 ft 0.0 log ç çD è 84 æ R = 1.122) = 0.1 = 2.15) (3.90 = 0.16 + 2.22) Based on the range of n-values indicated above a n-value of 0.167 = 0.0342 (0.62 ft ) 3.020 (3.019 æ 5.342 D.0927 (5. n = 0.0927 y 1 0 æ y0 1.18) (Assume yo = 5.25 (3.0 ö ÷ ÷ + 1.388 D .0395 (0.018 167 n = 0.162 n = 0.00 ö ÷ ÷ = 1.0 logç ÷ è 0.018 is selected.037 50 with D 50 in ft.0144 ÷ è ø ø = 8.167 = 0.16) /6 1/ 6 n = 0.0131).20 (3.167 = 0.342 D1 = 0.395 D1 = 0.0927 R 1/ 6 f 1/ 2 = 0.22) (3.032 50 with D 50 in ft.038 (0.20) f1/2 = 0.92 ft 0.1 = 6.62 ft 0.21) æ R = 2.0342 ) (0. the maximum and minimum n value could be used and a decision made based of the velocity and discharge.

(b) For the gravel sized material with the same values of τo = 0.095/ (165 .94)1/2 (./6 1/ 6 n = 0.0008 ft 0.095/ (165 .0948 lb/ft2.0 ft x 0. Critical Velocity for Beginning of Bed Material Movement (a) Sand Size Bed Material Given depth y is 12 ft and bed material size D50 is 0.69 This point plots above the incipient motion line in Shields (Figure 3. what is the critical velocity Vc? Vc = 11.γ) Ds = 0.25 y 1/ 6 D1/ 3 = 11.095/1.0015 ft 0.24mm = 0.25 x 121/ 6 x 0.0102) = 0.17) (3.037 90 with D 75 in ft.00102) = 0.92 V* Ds/ν = (0.62.00102 1/ 3 = 1.4 lb/ft3 x 4.038 (0. Ds = 3. τo/(γs . Recall that τo = γRS and V* = τ o / ρ .00038 = 0.41 x 10-5 = 15.00038 ft/ft 0.4) (0.0 ft 0.46mm = 0.18) 3.31 mm = 0.388 D1 = 0.92 ft ) (3. is 62.15) and indicates this channel bed is in motion.0102 ft we have: τo/(γs .62.8.4) (0.75 ft ) /6 1/ 6 n = 0.037 75 with D 75 in ft.091 V* Ds/ ν= (0.038 D1 = 0.4 PROBLEM 4 Beginning of Motion Bed Material Movement Using Shields Figure To establish if the bed material of a channel is in motion the Shields' relationship can be used.00102 ft.41 x 10-5 = 160 This point plots above the incipient motion line in Shields (Figure 3.72 ft / s 3. n = 0.0014 ft The shear stress on the bed.0388 (0. (a) For the following conditions determine if the bed material in a sand bed channel is in motion. R S D50 D16 D90 D84 = = = = = = 4.00102 ft 0. n = 0.095/1.1 mm = 0.15) and indicates the bed of the channel is in motion.095 lb/ft2.42mm = 0.001)/ 1.64 .0102)/ 1.γ) Ds = 0. at a single vertical in the cross-section.31mm = 0.94)1/2 (0.

104 ft 1/ 2 121/ 2 y (31.0007 V 3 (0.0 3 = = 0.0 3 = = 0. Determine the critical bed material size D50.0 ft/s Dc = 0.25 y 1/ 6 D1/ 3 = 11. 8.0007 ) 8.349 ft 1/ 2 121/ 2 y (106 mm) V = 16 ft/s Dc = 0.25 y 1/ 6 D1/ 3 = 11.0 ft/s. and 16 ft/s.421/ 3 = 12.1 mm = 0.0 3 = = 0.0 ft/s. V = 2.0 ft/s Dc = 0.0007 ) 2.25 121/ 6 x 0.0129 ft 1/ 2 121/ 2 y (3. 12 ft/s.0007 V 3 (0. what is the critical velocity Vc? Vc = 11.00162 ft 1/ 2 121/ 2 y (0. respectively.49 mm) V = 4.0007 ) 16.0007 V 3 (0. 4 ft/s.75 ft / s Critical Size for Beginning of Bed Material Movement Given depth y is 12 ft and velocity V is 2.70 ft / s (c) Cobble size Bed Material Given depth y is 12 ft and bed material size D50 is 128 mm = 0.0 3 = = 0.0102 ft.0 3 = = 0.65 .25 x 121/ 6 x 0.0102 1/ 3 = 3.(b) Gravel Size Bed Material Given depth y is 12 ft and bed material size D50 Is 3. what is the critical velocity Vc? Vc = 11.0007 ) 4.0007 ) 12.7 mm) V = 12 ft/s Dc = 0.828 ft 1/ 2 121/ 2 y (252 mm) 3.0007 V 3 (0.93 mm) V = 8 ft/s Dc = 0.42 ft.0007 V 3 (0.

52 ft V =16 ft/s y= 4. 5 2 = 0.0 6 0. 5 2 = 0.94 x 10 −7 V 6 D2 = ( 4.94 x 10 −7 ) 8.94 x 10 −7 ) 16.305 mm = 0.Critical Depth When the Bed Material Movement Would Stop (a) Given velocity V of 2. V = 8 ft/s y= 4.001 ft.0 6 0. V = 2 ft/s y= 4.66 .0001 ft There would not be any bed material movement at this velocity.5 2 = 33. 8.94 x 10 −7 ) 2.0 ft/s and bed material size D50 of 0.6 ft (b) Given velocity V of 2.0 6 0.0 6 0.0012 = 31.94 x 10 −7 V 6 D2 = ( 4.28 ft Would this depth occur? No.5 ft.0 and 16 ft/s and bed material size D50 of 152 mm = 0.94 x 10 −7 V 6 D2 = ( 4. determine the critical depth y when bed material movement would stop. y= 4. 3. Why not? The coarser material would armor the bed.94 x 10 −7 ) 2. determine the critical depth y for no bed material movement.94 x 10 −7 V 6 D2 = ( 4.0.

a stepwise application procedure for sediment transport calculations is outlined. the first two sections deal with definitions and general concepts. Bed material discharge (load): That part of the total sediment discharge which is composed of grain sizes found in the bed.1 INTRODUCTION This chapter describes key terms and several methods of computing sediment transport in alluvial channels. immediately above the bed. 4. several grain diameters thick (usually taken as two grain diameters thick). and provides references to those equations not discussed in detail in this chapter. The chapter also includes the derivation of the basic suspended bed sediment transport equation to illustrate the significant physical processes of sediment entrainment into the flow. The Yang sand and gravel total load equations are also suggested as a basic approach for hand calculation.e. Appendix B includes the results of testing several widely used sediment transport equations using a large compilation of river data. Bed material: The sediment mixture of which the stream bed is composed. contact load.1 . Finally. Bed layer: The flow layer. 4. The power function equations are well suited for quick estimates of sediment transport capacity and are easily adapted to field measurements. In addition. The total transported bed material discharge (or bed sediment discharge) is assumed equal to the transport capacity of the flow. The overview suggests the range of applicability of these equations (based on bed-material size). Simple sediment transport equations. bed load (contact load) plus suspended bed material discharge (load). Numerous sediment transport formulae have been developed with a wide range of laboratory and field conditions. Bed load: Sediment that moves by rolling or sliding along the bed and is essentially in contact with the stream bed in the bed layer. Since a significant portion of understanding sediment transport processes centers on correct use of the terminology. are also presented in this chapter. an overview of selected sediment transport equations is presented. i. in the form of power functions.CHAPTER 4 SEDIMENT TRANSPORT 4.e. Three classic sediment transport formulae are then discussed to illustrate the application of sediment transport theory.2 DEFINITIONS In this section the basic terms for describing sediment load in alluvial channels are summarized. i.

2 . or the ratio of the dry weight of sediment to the total weight of water-sediment mixture in a sample or a unit volume of the mixture. Also called fine sediment discharge (load). It is the sum of the suspended-sediment discharge and the contact sediment discharge. Sediment concentration (by weight or by volume): The quantity of sediment relative to the quantity of transporting fluid.Contact load: Sediment particles that roll or slide along in almost continuous contact with the stream bed. suspended in. Suspended load (or suspended sediment): Sediment that is supported by the upward components of turbulence in a stream and that stays in suspension for an appreciable length of time. The concentration may be by weight or by volume.e. or fluid-sediment mixture. Sediment yield: The dry weight of sediment per unit volume of water-sediment mixture in place. The mass per unit volume including both water and Discharge-weighed concentration: The dry weight of sediment in a unit volume of stream discharge. the concentration is always in ratio by weight. Sediment (or fluvial sediment): Fragmentary material that originates from weathering of rocks and is transported by. Sediment discharge: The quantity of sediment that is carried past any cross-section of a stream in a unit of time. Density of water-sediment mixture: sediment. When expressed in ppm or mg/l. Suspended bed load: The portion of suspended load which is composed of grain sizes found in the bed. or the sum of the bed sediment discharge and washload. Suspended-sediment discharge: The quantity of suspended sediment passing through a stream cross-section outside the bed layer in a unit of time. Washload: That part of the total sediment discharge (load) which is composed of particle sizes finer than those found in appreciable quantities in the bed and is determined by available bank and upstream supply rate.. Unmeasured sediment discharge: Sediment discharge close to the bed that is not sampled by a suspended sediment sampler. or deposited from water. Total sediment discharge: The total sediment discharge of a stream. or the sum of the measured sediment discharge and the unmeasured sediment discharge. Load (or sediment load): The sediment that is being moved by a stream. or the ratio of the discharge of dry weight of sediment to the discharge by weight of water-sediment mixture normally reported in parts per million (ppm) or parts per liter (ppl). 4. bed load. i.

velocity distribution. The coarser part of the load. distribution. i. not only between streams but at a given point of a single stream. Group 1 . viscosity and density of the fluid sediment mixture. is limited by its availability in the watershed. magnitude. Each of these two conditions may limit the sediment rate at the cross-section. hydraulic radius. the two sources of sediment transported by a stream are: (1) bed material that makes up the stream bed. The variables which control the amount of sediment brought down to the stream are subject to so much variation. Group 2 – The capacity of a stream to transport sediment depends on hydraulic properties of the stream channel. It is practicable. 4.. roughness. and season of rainfall. A list of these variables is given as follows.3 . cultivation and grazing. surface erosion and bank cutting. the part that is more difficult to move by flowing water. turbulence. depending on the relative magnitude of two controls: the availability of the material in the watershed and the transporting ability of the stream. duration. intensity. the part which the flow can easily carry in large quantities. The variables that deal with the capacity of the stream to transport solids are subject to mathematical analysis. tractive force. soil moisture conditions. This part of the load is designated as washload.1 Source of Sediment Transport Einstein (1964) stated that: Every sediment particle which passes a particular cross-section of the stream must satisfy the following two conditions: (1) it must have been eroded somewhere in the watershed above the cross-section. is limited in its rate by the transporting ability of the flow between the source and the section. in some cases. and from these records to determine a soil loss from the area. and size and gradation of the sediment. In most streams.Sediment brought down to the stream depends on the geology and topography of watershed. vegetal cover.e. their effect is not definitely known. discharge. to measure the sediment discharge over a long period of time and record the results. the finer part of the load.3 GENERAL CONSIDERATIONS The amount of material transported or deposited in the stream under a given set of conditions is the result of the interaction of two groups of variables. that the quantitative analysis of any particular case is extremely difficult. i. velocity. In the first group are those variables that influence the quantity and quality of the sediment brought down to that section of the stream. In the second group are variables that influence the capacity of the stream to transport that sediment. however. These are fluid properties. Thus. Geologically both materials come from the watershed. and (2) fine material that comes from the banks and the watershed (washload). slope. for engineering purposes.4.e. These variables are not all independent and. But for 4. These variables are closely related to the hydraulic variables controlling the capacity of the stream to carry water.. (2) it must be transported by the flow from the place of erosion to the cross-section.3. This part of the load is designated as bed sediment load.

4 .062 mm which is also the division point between sand and silt. Instead. Even as there is no sharp demarcation between bed material discharge and washload there is no sharp line between contact load and suspended sediment load. For these conditions. however. By method of measurement LT = Lm + Lu (4. The sediment discharge consisting of grain sizes smaller than 0.1 (Julien 1995): 1.062 mm is considered as washload. As a rule of thumb. By source of sediment L T = L w + L bm (4. if the bed of a channel is silt. that give rise to non-equilibrium conditions. The washload is not transported at the capacity of the stream. The sediment transport capacity of silts has been addressed in some transport equations. the rate of bed load transport of the silt sizes is less a question of supply than of capacity.the engineer. including a pictorial representation of measured load and unmeasured load follows. 4. When a river reaches equilibrium. In fact.062 mm for coarse bed sediment streams. It is important to note that in a fast flowing mountain stream with a bed of cobbles the washload may consist of coarse sand sizes. most rivers are subject to some kind of control or disturbance. its transport capacities for water and sediment are in balance with the rates supplied. The equations for estimating the total bed material discharge of a stream are based on these laws. the washload depends on availability and is not functionally related to measurable hydraulic variables. natural or human-induced. In contrast. No sharp demarcation exists between washload discharge and bed material discharge.1a) 2. A particle may move part of the time in contact with the bed and at other times be suspended by the flow.2 Mode of Sediment Transport Sediment particles are transported by rolling or sliding on the bed (bed load or contact load) or by suspension by the turbulence of the stream. the criterion based on D10 might still be applicable. the transport of sand sizes is supply limited.1c) 4. Total sediment load can be expressed by three equations as illustrated in Figure 4. A further subdivision of mode of transport of sediment. This invalidates the criterion based on Ds < 0. The distinction is important because the two modes of transport follow different laws. A more reasonable criterion is to choose a sediment size finer than the smallest 10 percent of the bed material as the dividing size between washload and bed sediment load.1b) 3. By type of movement L T = Lb + L s (4. many engineers assume that the bed material discharge is composed of sizes (Ds) equal to or greater than 0. the distinction is important because the bed material is transported at the capacity of the stream and is functionally related to measurable hydraulic variables.3.

The contact load is not measured. or the sum of the contact sediment discharge (bed load) and suspended sediment discharge. In the former sum. the measured suspended-sediment discharge is from 90 to 95 percent of the total sediment discharge.where: LT Lb Ls Lm Lu Lw Lbm = = = = = = = Total load Bed load which is defined as the transport of sediment particles that are close to or maintain contact with the bed Suspended load defined as the suspended sediment passing through a stream cross-section above the bed layer Measured sediment Unmeasured sediment that is the sum of bed load and a fraction of suspended load below the lowest sampling elevation Wash load which is the fine particles not found in the bed material (Ds < D10). part of the suspended sediment in a vertical is not measured. The sediment load that is measured by suspended-sediment samplers consists of both the washload (fine sediment load) and suspended-sediment load. the total sediment discharge is based on source of the sediments and the latter sum is based on the mode of sediment transport. and because suspended sediment samplers cannot travel the total distance in the vertical to the bed. 4. 4. However. Whereas suspended sediment load consists of both bed sediments and fine sediments (washload). in shallow sand-bed streams with little or no washload the measured suspended-sediment load may be as small as 50 percent of the total load. and originates from available bank and upstream supply Capacity limited bed material load Figure 4. the amount of bed material moving in contact with the bed of a large sand-bed river is from 5 to 10 percent of the bed material moving in suspension.3.3 Total Sediment Discharge The total sediment discharge of a stream is the sum of the bed sediment discharge (bed material load) and the fine sediment discharge (washload). Classification of sediment transport in streams (rivers). only the bed sediment discharge can be estimated by the various equations that have been developed.2). The presence of high concentrations of wash load affects the apparent viscosity of the water-sediment mixture and reduces the fall velocity of silt and sand grains.1. In general. Generally. The fine sediment discharge (washload) depends on its availability not on the transporting capacity of the flow and must be measured (Figure 4.5 .

The amount of material transported in suspension varies with depth. For the same water discharge. with the highest concentrations occurring near the bed. the difficulty of measuring many of the variables and the statistical nature of bed material transport. At equilibrium transport. The variation between the magnitude of the bed sediment discharge predicted by different equations under the same conditions can be significant. Julien 1995).6 . Many equations have been developed for the estimation of bed sediment transport. Suspended-sediment concentrations as large as 600. the washload can be eliminated from the measured suspended-sediment load if the size distribution of the material is known. This can be expected given the number of variables. is important. useful bed material discharge information can be obtained. Concentrations of this magnitude are largely fine sediments.4 SUSPENDED BED SEDIMENT DISCHARGE The sediment transport process is best described through a discussion of suspended bed sediment discharge. settling is small in comparison to turbulent mixing and the sediment concentration is vertically uniform. knowledge of the river.000 ppm or 60 percent by weight have been observed. the fine material in the flow increases the capacity of the flow to transport bed material.Figure 4. By increasing fluid properties (viscosity and density). with proper use. and knowledge of the limitations of each method. which usually accounts for the majority of the total load. both in suspension and in contact with the bed. For very fine particles. 4.2. The sediment load of a stream at a cross section or through a reach of a stream can be determined by measuring the suspended-sediment portion of the load using samplers and estimating the unmeasured discharge or by using one of the many methods that have been developed for computing the bed sediment load and estimating the washload. the predicted sediment discharge can have a 100-fold difference between the smallest and the largest value. the interrelationships among them. The magnitude of the suspended or bed sediment discharge can be very large. In these cases. only the bed sediment load. the vertical exchange of sediment is balanced between particle settling (due to gravity) and turbulence (that mixes higher concentration flow up into the water column). Sediment transport capacity and supply curves (Shen 1971. Nevertheless. In many problems. Simons and Sentürk 1992. For 4.

4. is obtained by integrating Equation 4.7 . Normally this interrelation is neglected for low sediment concentrations or a coefficient applied to compensate for it. Schematic sediment and velocity profiles.3. and γs is the weight per unit volume of the suspended sediment.2) where v and c vary with y and are the time-averaged flow velocity and volumetric concentrations.3) where: C = Average suspended-sediment concentration by volume The vertical distribution of both the velocity and the concentration vary with the mean velocity of the flow. The suspended bed sediment discharge in Newtons or lbs per second per unit width of channel." The level "a" is generally assumed to be 2-grain diameters above the bed layer. The discharge of suspended sediment for the entire stream cross-section.3. The one-dimensional gradient type diffusion equation is employed to obtain the vertical distribution for c and the logarithm velocity distribution is assumed for v in turbulent flows. Also v and c are interrelated.coarser particles. QS. To integrate Equation 4.2 over the cross section to give Q S = γ S QC (4. respectively. bed roughness and size of bed material. v and c must be expressed as functions of y. the velocity and turbulence at a point is affected by the sediment at the point.2. uniform two-dimensional flow is yo qS = γ s ò a vcdy (4. for steady. The distributions are illustrated in Figure 4. and the sediment concentration at the point is affected by the point velocity. settling is rapid and particles that are in suspension tend to concentrate nearer the bed. That is. qs. Figure 4. The integration is taken over the depth between the distance "a" above the bed and the surface of the flow "yo. Sediment movement below this level is considered bed load rather than suspended load.

7) τ ρ (dv / dy )) where: τ and dv/dy = Shear stress and velocity gradient. To obtain an expression for εs the assumption is made that ε s = βε m (4. In order to determine the value of c at a given y. which characterizes the magnitude of the exchange of particles across any arbitrary boundary by the turbulence = Concentration gradient = Average rate of settling of the sediment particles = Average rate of upward sediment flow by diffusion Integrating Equation 4. at point y For two-dimensional steady uniform flow τ = γS( y o − y ) = τ o (1 − y / y o ) (4. respectively. also called the mass transfer coefficient.The one-dimensional diffusion equation describes the equilibrium condition when the quantity of sediment settling across a unit area due to the force of gravity is equal to the quantity of sediment transported upwards resulting from the vertical component of turbulence and the concentration gradient. the value of ca and the variation of εs with y must be known.6) where: εm εm = = Kinematic eddy viscosity or the momentum exchange coefficient defined by (4.4) where: ω c εS dc/dy ωc εS dc/dy = Fall velocity of the sediment particle at a point = Concentration of particles at elevation y above the bed = Exchange coefficient.5) where ca is the concentration of sediment with settling velocity equal to ω at a level y = a.8 . The resulting equation for a given particle size is ωc = − ε S dc / dy (4.4 yields yo c = c a exp − ω ò a dy / ε s { } (4.8) 4.

10) where: β κ V* = = = Coefficient relating εs to εm Von Karman's velocity coefficient assumed equal to 0.12. The value of z is small for large shear velocities V* or small fall velocities ω. shows that for small values.9 . β is approximately 1. An evaluation of the Rouse number. The substitution of Equation 4. z. the equation for suspended sediment transport becomes 4. For fine particles.10 into Equation 4. the concentration profiles are uniform.4 shows a family of curves obtained by plotting Equation 4. named after the engineer who developed the equation in 1937 Figure 4.10 indicates that εm and εs are zero at the bed and at the water surface. Using the logarithmic velocity distribution for steady uniform flow and Equation 4.74 dv = dy τo / ρ κy = V* κy (4. for small particles or for extremely turbulent flows. The values of β and κ have been investigated. Thus.12) where: c ca z = = = Concentration at a distance y from the bed Concentration at a point a above the bed ω/βκV*. For large z values. little sediment is found near the water surface. and have a maximum value at mid-depth.12 for different values of the Rouse number z.4 Shear velocity equal to gRS in steady uniform flow Equation 4. the sediment distribution is nearly uniform.4 gives dc −ω = c βκV* dy y(1 − y ) yo (4.4. but apparently decreases with increasing sediment concentration. Also. it is well known that in clear water κ = 0.and from Equation 2.0.9) Thus ε s = βε m = βκV* y(1 − y / y o ) (4.11) and after integration éy − y a ù c =ê o ú ca ë y yo − a û Z (4. the Rouse number.

2 ê ú ê2.12 (SI) and 4.7.10 .5 BED SEDIMENT DISCHARGE In this section.13 (English).5.8) because of their frequent application and wide acceptance. and Colby (1964) methodologies.5.2. An example calculation for a suspended sediment concentration profile is given in Section 4. gravel.5.5.4.4) methods are bed material load equations used for sand-bed streams.13) Many investigators have solved this equation through integration. The Einstein (Section 4.2) and Colby (Section 4. lignite. and barite: æ Qb ç ç Q è öæ K B ÷ ÷ç ç øè K r ö ÷ ÷ ø 3/2 æγö γy o S f = B ′(γ s − γ )D m + B ç ç g÷ ÷ è ø 1/ 3 æ γs − γ ö ç ÷ ç γ ÷ è s ø 2/3 qB 2/3 (4. Graph of suspended sediment distribution (Rouse 1937). These are the Meyer-Peter and Müller (1948). The assumptions and integration made by Einstein are presented in Section 4.5. and cobbles. three classic sediment transport formulae are discussed in detail to illustrate sediment transport processes. Figure 4. 4.5 lnç ç y û ê kS ÷ ú øû ë yo − a è ë Z (4. Einstein (1950). natural gravel. The Yang sand and gravel total load equations are also included (Section 4.14) 4. 4. power function sediment transport relationships are discussed to provide a practical method for quick sediment transport calculations. In Section 4.q S = γV* c a ò yo a é a æ yo − y ù é Xy ö ù ÷ ú dy 30.1) is applicable to streams with bed material consisting of sand. sand particles of mixed sizes and density.1 Meyer-Peter and Müller Bed Load Equation Meyer-Peter and Müller (1948) developed the following bed load equation based on experiments with sand particles of uniform sizes. The Meyer-Peter and Müller bedload equation (Section 4.

This equation is é æ Qb qB = K u1 êK u2 ç ç Q ê è ë 1/ 6 öæ ç D 90 ÷ ÷ç n øè b ö ÷ ÷ ø 3/2 ù y o S f − 0.18) where D90 is in meters. If the boundary is hydraulically rough. Bureau of Reclamation in 1960 and is expanded here for application in SI units.17) Where fb′.25 for sediment transport and is meaningless for no transport since qB is zero and the last term drops out.047 for sediment transport and 0. B has a value of 0.11 . Equation 4. Kr is given by Kr = 26 D 90 1/ 6 (4. B = = = = = = Bed-load rate in weight per unit time and per unit width Water discharge quantity determining bed load transport Total water discharge Depth of flow Energy slope Dimensionless constants B′ has the value 0.19) 4.15) and Kr = V R 2/3 S ′f 1/ 2 (4. The quantities KB and Kr are defined by the expressions KB = V /2 R S1 f 2/3 (4. (V*D90/ν≥ 100). the Darcy-Weisbach bed friction factor for the grain roughness.S.034 for the case of no sediment transport.14 is dimensionally homogeneous so that any consistent set of units may be used.Sf′ Therefore KB = Kr ′ fb 8 V gRS f = = = Total energy slope Part of the total slope required to overcome grain resistance Part of the total slope required to overcome form resistance (4.627D m ú ú û 3/2 (4.14 was converted to units generally used in the United States in the field of sedimentation for water and quartz particles by the U.where: qB Qb Q yo Sf B′. Equation 4.16) where: Sf Sf′ Sf . fb′ is determined from the Nikuradse pipe friction data with pipe diameter equal to four times the hydraulic radius and Ks = D90.

where: Ku1 Ku1 Ku2 Ku2 qB Qb Q D90. respectively Horizontal side slope related to one unit vertically Bottom width The ratio Qb/Q for rectangular channels is given by Qb = Q 1 æ 2y 1+ ç o è W öæ n w ÷ç ç øè n b ö ÷ ÷ ø 3/2 (4.20) where: pi = Percentage by weight of that fraction of the bed material with geometric mean size. m3/s (cfs) Total water discharge m3/s (cfs) In millimeters (both SI and English units) The quantity Dm is the effective diameter of the sediment given by Dm = Σ ip iD si 100 (4.780 in SI units 1.22) where: n.306 in English units Metric-tons/day/meter (Tons/day/foot) Water discharge quantity determining the bed-load transport. of the bed.606 in English units 10.846 in SI units 3. nw Hs W = = = Roughness coefficients of the total stream.23) 4.12 . Dm = = = = = = = = 4.21) and for trapezoidal channels 3/2 2 1/ 2 ì ï 2y o (1 + Hs ) é æ n w ö ù ü ï n b = ní1 + ÷ úý ê1 − ç W ê è n ø û úï ï ë î þ 2/3 (4. Dsi The term nb is the Manning's roughness coefficient for the bed of rectangular channels é 2y n b = nê1 + o W ê ë æ æ n w ö 3 / 2 öù ç1 − ç ÷ ÷ú ç è n ø ÷ú øû è 2/3 (4. and of the banks. nb.

5. there is no sharp demarcation between the contact bed sediment load and the suspended bed sediment.4.and for trapezoidal channels is Qb = Q 1 1/ 2 2y (1 + H2 s) 1+ o W æ nw ç çn è b ö ÷ ÷ ø 3/2 (4. the contact load is given as 4. Einstein's bed sediment discharge function gives the rate at which flow of any magnitude in a given channel transports the individual sediment sizes which make up the bed material. This makes his equations extremely valuable where it is necessary to determine the change in bed material with time. this division is warranted by the fact that there is a difference in behavior of the two different loads which justifies two physical equations.24) The Meyer-Peter and Müller formula (Equation 4.14) is often written in the form qb = K ( τ − τ c ) 3 / 2 (4.2 Einstein's Method of Computing Bed Sediment Discharge Einstein’s (1950) method is included in detail because the formulation provides an excellent discussion of sediment transport processes. The method also incorporates the vertical velocity distribution based on the roughness of the boundary.26b) (4. 4.13 (English). Each size moves at its own rate.25) where: ù é ú ê 1 ú ê K=ê 2/3 ú 1/ 3 γ ö æ γs − γ ö ú ê Bæ ÷ ÷ ç ç ç ê g÷ ç γ ÷ ú ë è ø è s ø û æ Qb τ=ç ç Q è öæ K B ÷ ÷ç ç øè K r ö ÷ ÷ ø 3/2 3/2 ≅ 12. However. The total bed sediment discharge is the sum of the contact load and the suspended load. The method computes a contact load concentration and uses this concentration as the starting point for integrating the suspended load as presented in Section 4. As mentioned earlier.26a) γy o S f (4.12 (SI) and 4.13 .26c) τ c = B ′(γ s − γ ) D m An example of sediment transport calculations using the Meyer-Peter and Müller equation is given in Section 4.9 γs ρ (4. The equations describe and incorporate the physical processes of contact load and suspended load. For each size Ds of the bed material.

33) and γs ρ ρs g φ* = = = = = Unit weight of sediment Density of the water Density of the sediment Gravitational acceleration Dimensionless sediment transport function = f(ψ*) given in Figure 4. suspended and contact bed sediment discharges qT.34) (4.30) where iT.27) and the suspended sediment load is given by is qs (4.29 becomes i T q T = iB qB (1 + PEI1 + I2 ) (4.3 log (30. With suspended sediment load related to the contact load.29) and finally Q T = Σi T q T (4. Equation 4. qs and qB for a given grain size Ds.32) I1 and I2 are integrals of Einstein's form of the suspended sediment Equation 4.6 / β x ) 2 ψ æ ρs − ρ ö Ds ψ′ = ç ç ρ ÷ ÷ R′ S è ø b f ξ = Correction factor given as a function of D / X in Figure 4. and iB are the fractions of the total. The suspended sediment load is related to the contact load because there is a continuous exchange of particles between the two modes of transport.iB qB (4.6 s 4.2y o / ∆ ) (4. is.5 (4. The term QT is the total bed sediment transport.12 iB qB = φ * iB γ s æ ρ 1 ö ç ÷ ç ρ − ρ gD 3 ÷ s ø è s 1/ 2 (4.35) ψ * = ξY(β / β x ) 2 ψ = ξY(log10.14 .31) where: PE = 2.28) and the total bed material discharge is i T q T = i s q s + i B qB (4.

36a) 01.216 (1 − E) Z E Z −1 (1 − E ) Z ò 1 E é1 − y ù ê ú dy ë y û é1 − y ù ê ú In ydy ë y û Z Z (4.38a) (4.4 V* Fall velocity of the particle of size Ds Ratio of bed layer thickness to flow depth.216 ò 1 E (4.40) where: Z ω E yo a = = = = = ′ ω / β κV*′ = ω / 0.8 = log (10.7 = 11.9 and 4. 2D65 (4.8 (4.8 (4. if ∆/δ′ > 1.39δ′.77∆. a/yo Depth of flow Thickness of the bed layer.6 X / ∆ ) E Z −1 I1 = 0. Figure 4.75 log (30. if ∆/δ′ < 1. 4. Einstein's φ * vs ψ * bed load function (Einstein 1950).15 .∆ X X X = = = = δ′ ′ v / V* V*′ R′ b ′ ′ Rb Sf 0.5.2y / ∆ ) (4.36b) Apparent roughness of the bed.39) I2 = 0.41) The two integrals I1 and I2 are given in Figures 4.37) = Einstein’s velocity distribution equation = 5.6ν / V*′ (4. k s / X Correction factor in the logarithmic velocity distribution equation given as a function of k s / δ′ in Figure 4.38b) Y βx = Shear velocity due to grain roughness = gR ′ S b = Hydraulic radius of the bed due to grain roughness = R b − R ′ ′ b = Hydraulic radius of the bed due to channel irregularities = Slope of the energy grade line normally taken as the slope of the water surface = Another correction term given as a function of D 65 / δ′ in Figure 4.10 as a function of z and E.

6. Figure 4.7. Hiding factor (Einstein 1950).16 . 4.Figure 4. Einstein’s multiplication factor X in the logarithmic velocity equations (Einstein 1950).

75 log(12.26R ′ b / k s X) (4. The part of energy corresponding to grain roughness is transformed into turbulence that stays at least for a short time in the immediate vicinity of the grains and has a great effect on the bed load motion.42a) or V / V*′ = 5. the shear velocity is based on the hydraulic radius of the bed due to grain roughness Rb′. Its computation is explained in the following paragraph.75 log(12. The transmission of shear to the boundary is accompanied by a transformation of flow energy into energy turbulence.42b) 4. This energy does not contribute to the bed load motion of the particles and may be largely neglected in the sediment transportation.26R ′ b / ∆) (4.Figure 4. Einstein's equation for mean flow velocity V in terms of V*′ is V / V*′ = 5.17 .8. surface drag and form drag. or at a considerable distance away from the grains. the other part of the energy which corresponds to the form resistance is transformed into turbulence at the interface between wake and free stream flow. Pressure correction (Einstein 1950). Total resistance to flow is composed of two parts. In the preceding calculations for the bed sediment load. Whereas.

Integral I1 in terms of E and Z (Einstein 1950). 4.9.18 .Figure 4.

10.19 .Figure 4. 4. Integral I2 in terms of E and Z (Einstein 1950).

slope S. The procedure for computing total bed sediment discharge in terms of different size fractions of the bed material is: (1) (2) (3) (4) (5) Calculate ψ* using Equation 4.44 is given in Figure 4.44) The relation for Equation 4.42 and Figure 4.43) where ψ′ = ρ s − ρ D 35 ρ R′ bS f (4.11. Friction loss due to channel irregularities.32 Sum up the qB across the flow to obtain iBQb Sum up the size fractions to obtain Qb 4.20 . Figure 4. hydraulic radius Rb and the size of bed material are known. If mean velocity V.Furthermore. then Rb′ is computed by trial and error using Equation 4.11. The procedure to follow in computing Rb′ depends on the information available. as a function of sediment transport (after Einstein and Barbarossa 1952).11. Einstein suggested that V / V*′′ = θ[ψ ′] (4.5 for each size fraction Calculate iBqB for each size fraction using Equation 4.33 for each size fraction Find φ* from Figure 4.

This supports the premise that the Meyer-Peter and Müller equation is most applicable to coarse grain sizes with little or no suspended load.9 and 4.3 Comparison of Meyer-Peter and Müller and Einstein Contact Load Equations Chien (1954) has shown that the Meyer-Peter and Müller.21 .12.10 Calculate PE using Equation 4.12 shows the comparison of Equation 4.45 with Einstein's ψ* vs.31 Sum up all the qB and all the iB to obtain the total suspended discharge Qss Thus. equation can be modified into the form æ 4 ö φ* = ç − 0.For the suspended sediment discharge: (6) (7) (8) (9) (10) (11) Calculate Z for each size fraction using Equation 4.13.41 Calculate E = 2Ds / yo for each fraction Determine l1 and l2 for each fraction from Figures 4. 4. φ* relation for uniform bed sediment size and for sediment mixtures using D35 in the Einstein relation and D50 in the Meyer-Peter and Müller relation. but diverge for fine sands.188 ÷ çψ ÷ è * ø 3/2 (4.45) Figure 4. the total bed sediment discharge is computed as: (12) Add the results of Step 5 and 11.32 Compute the suspended discharge from IB qB (1 + PE I1 + I2) from Equation 4. 4. They show good agreement for coarse sands. A sample problem showing the calculation of the total bed sediment discharge using Einstein's procedure is presented in Section 4. Comparison of the Meyer-Peter and Müller and Einstein methods for computing contact load (Chien 1954). Figure 4.5.

In applying Figures 4.4.13.14 for determining the bed sediment discharge. Required data are the mean velocity V. After investigating the effect of all the pertinent variables. four depths of flow. Figure 4.1 ft are extrapolated from limited data and theory. Colby’s method illustrates that washload can have a significant impact on bed material load transport capacity. the water temperature °F and the measured fine sediment concentration Cf.0 and 0. Colby (1964) developed four graphical relations shown in Figures 4. Colby was guided by the Einstein bed-load function (Einstein 1950) and a large amount of data from streams and flumes.14 to compute the bed sediment discharge. most curves of 10-foot depth and part of the curves of 1.13 and 4.5. the depth yo. Relation of discharge of sands to mean velocity for six median sizes of bed sands.13 and 4. the following procedure is used: 1. which can also be used to check the reasonableness of other sediment transport calculations. it should be understood that all curves for 100-foot depth. and a water temperature of 60°F (Colby 1964). Extrapolated curves are shown as dashed lines in Figure 4. In arriving at his curves. the median size of bed material D50.4 Colby's Method of Estimating Bed Sediment Discharge Colby’s (1964) method is a graphical method for estimating total load.13. One very important feature of the Colby method is the inclusion of a correction factor for fine sediment (washload) concentration effects. However.22 . 4.

Figure 4. The total sand discharge is Q T = Wq T (4. and D50 can be found from Figure 4. Two correction factors k1 and k2 shown in Figure 4. yo.47) where: W = Width of the stream Colby (1964) found that: "The agreement of computed and observed discharges of sands for sediment stations whose records were not used to define the graphs seemed to be about as 4. 3.14 is applied to correct for the effect of sediment size.46) As Figure 4.14 shows.2 mm and 0.3 mm.13 by first reading qn knowing V and D50 for two depths that bracket the desired depth and then interpolating on a logarithmic graph of depth versus qn to get the bed sediment discharge per unit width. the factor k3 from Figure 4.30 mm range. k1 = 1 when the temperature is 60°F. If the bed sediment size falls outside the 0. Unit bed sediment discharge qT corrected for the effect of water temperature.14. k2 = 1 when the concentration of fine sediment is negligible and k3 = 1 when D50 lies between 0. 4. 2.14 account for the effect of water temperature and fine suspended sediment on the bed sediment discharge.20 mm to 0. Uncorrected sediment discharge qn for the given V.23 . presence of fine suspended sediment and sediment size is given by the equation q T = [1 + (k 1k 2 − 1) k 3 ] qn (4. Colby's correction curves for temperature and fine sediment (Colby 1964).

and the relations can be modified to include the effect of washload and viscosity effects. Further. the median diameter of the bed sediment is maintained constant equal to 0. Pw = wetted perimeter of the banks. bed sediment size and depth on bed sediment and water discharge is examined in this section using Einstein's bed-load function (1950) and Colby's (1964) relationships. the kinematic viscosity was varied to correspond with variations in temperature from 39. 4. The effects of variables on sediment discharge were studied over approximately the same range of variation for each variable.030 mm.0 to 10. D35 = size of bed material for which 35 percent is finer.good as that for stations whose records were used. and 4." An example showing bed sediment discharge calculations by the Colby method is presented in Section 4.S. 4.2° to l00°F inclusive.18 shows the variation of the sediment discharge due to changes in the depth of flow for any given discharge. D50.17 and show the variation of sediment discharge due to changes in bed material size.15 shows that when the bed sediment becomes finer. The values of depth of flow varied from 1. Colby's relations are chosen because of the large amount and range of data used in their development.16).5. This indicates that the capacity of the stream to transport sands increases very fast for a small increase of discharge at constant depth.0 ft. The average water temperature was assumed to be equal to 70°F and the average energy gradient of the channel was assumed to be equal to 0. the curves for constant depth of flow show a steep slope.00095.3 m (270 ft). To study the relative influence of variables on bed material and water discharges.24 . Temperature is third in importance (Figure 4.13. iB = percentage of bed sediment in fraction i. In Figure 4. The channel width is also maintained constant at 82. and iB was accomplished by using the average bed material distribution given by Nordin (1964) and shifting the curve representing the average bed sediment distribution along a line parallel to the abscissa drawn through D50.000 ppm. New Mexico are used.7S to 1. 1940 to October 1.5 S . The water and sediment discharges were computed independently for each variation of the variables and for three subreaches of the Rio Grande of differing width near Bernalillo. slope. in which S is the average bed slope assumed to be equal to the average energy slope. A = cross-sectional area.17). The data required to compute the bed material discharge using Einstein's relations are: S = channel slope. 4. Geological Survey from October 1. The computed water and sediment discharges are plotted in Figures 4. the data taken by the U. Di = size of bed sediment in fraction i. The second most important variable affecting sediment discharge is the slope variation (Figure 4. slope and temperature for any given water discharge. Einstein's bed-load function is chosen because it is the most detailed and comprehensive treatment from the point of fluid mechanics.3 m (270 ft). Similar figures can be developed for other sizes of bed material. computed using Colby's (1964) relations. D65 = size of bed material for which 65 percent is finer. Di. and V = average velocity. the water temperature is assumed constant and the concentration of fine sediment is assumed less than 10.18. ν = kinematic viscosity. The variation of D65. γs = specific weight. Pb = wetted perimeter of the bed. The applicability of the results depends on the reliability of the modified Einstein bed-load function and Colby's relationships used in the analysis rather than on the choice of data.5 Relative Influence of Variables The study of the relative influence of viscosity. Figure 4. the sediment discharge increases considerably. Figure 4.16. In the analysis.15. 1970 on the Rio Grande near Bernalillo. The width of the channel reach was 82. nw = Manning's wall friction coefficient. the energy slope was varied from 0.

Bed-material size effects on bed material transport. Effect of slope on bed material transport.25 .Figure 4.16. Figure 4. 4.15.

4.18.Figure 4. Figure 4. Variation of bed material load with depth of flow.17. Effect of kinematic viscosity (temperature) on bed material transport.26 .

1 Introduction Power relationships relate sediment transport rates to hydraulic conditions and sediment characteristics. Their utility is in their ease of use and. medium to coarse sand. power relationships empirically relate sediment transport with hydraulic conditions and sediment characteristics. (1981) power function that was developed for steep sand and gravel bed channels.6.6. the Meyer-Peter and Müller bedload equation is still widely used. and the Colby method provides a graphical approach that can be used as a quick check on the reasonableness of the results of other sand-bed sediment transport calculations. 2001) includes a power function equation developed using the Yang (1996) sand transport equation. three classic sediment transport equations were presented. The resulting equation is site specific. although. and particle diameter. The value of power relationships is in (1) their ease of use. and (3) application to site-specific conditions.2 Basic Power Function Relationship In 1981. is also presented. HEC-20 (Chapter 6) illustrates the application of power relationships to predicting the equilibrium slope for a channel that is either sediment deficient or has an excessive sediment supply. The following sections present the Simons et al. 4. velocity. (1981) power function was developed specifically for supercritical flow conditions on sand and fine gravel bed channels. Power relationships can be developed by computing sediment transport with a more rigorous approach and fitting the results to a power function form. A more general power function relationship. but can be used. The Kodoatie equation can be used for silt. with judgment. HEC-20 (Lagasse et al. fine sand. (1999).6 POWER FUNCTION RELATIONSHIPS In Section 4.48) 4. when developed from measured data. by Kodoatie et al. and gravel bed rivers by selecting appropriate coefficients and exponents. as indicated above. Power relationships are also well suited for correlating measured sediment transport with observed hydraulic conditions. and gradation coefficient. they can be integrated into computer solutions. To provide additional basic approaches. The Simons et al. These provide a practical method for quick sediment transport calculations and are readily adaptable to computer solutions.5. proposed an efficient method of evaluating sediment discharge.27 . Simons et al. 4. In summary. Of the three. Most power function relationships correlate sediment transport to velocity and flow depth. (2) application to specific flow regimes. their sitespecific accuracy. power function relationships are discussed in this section. to estimate sediment transport rates for hydraulic conditions outside the range of observations. They can be developed by fitting the coefficients to computed sediment transport data from more sophisticated equations or to measured data. This simple relationship can be used as a first estimate of sediment transport capacity or as a comparison with the more rigorous equations. The basic power function relationship is: q s = c s1 y c s 2 V c s 3 (4. The method is based on the variables flow depth. Both are simple enough in their application to be considered "basic" approaches for hand calculation.4.

This is the only transport relationship specifically developed for upper flow regime conditions.64 x10 6. from a computer solution of the Meyer-Peter and Müller bed load transport equation and the integration of the Einstein method for suspended bed material discharge (Julien 1995).).where: qs cs2.93 3.24 -0. These power relationships were developed by Simons et al.61 3.95 3.91 3. Table 4.36 cs2 3.9 -6 -6 9.18 -0.06 -0.55 3.27 3.78 x10 0.60 x10 0.3048(2 . The term Gr in Table 4.94 x10 0.09 -0.21 3. Coefficient and Exponents of Equation 4.05 x10 0. respectively. CS1 needs to be multiplied by a factor of 0. n = 84.82 cs3 -5 -6 Gr = 2 cs1 1.1 is defined as the gradation coefficient of the bed material and is: Gr = 1 é D 50 D 84 ù + ê ú 2 ë D16 D 50 û (4. ft/s (m/s) This relationship can be applied in steep streams with sand and fine gravel beds that normally exhibit critical or supercritical flow.25 0.12 -0. Table 4.0 -6 -6 -6 -6 -6 5.92 3.28 .3048 ( 2 − C s 2 −C s 3 ) (c s1 ) 4.3 3.Cs2 . ft (m) Mean flow velocity.76 3.91 3. and 16.80 x10 0.86 -5 1.5 D50 (mm) 1.3 -0.10 x10 7.66 cs3 Gr = 4 cs1 cs2 Cs3 * c s1 (SI units) = 0.71 -5 -5 -6 cs1 3.0 3.51 0.715 0.28 cs2 3.Cs3).44 x10 0.48 (Simons et al.14 x10 7.62 x10 5. cs3 cs1 y V = = = = = Unit sediment transport rate.495 0.94 x10 6.34 3.62 x10 6. In this case.1 Gr = 1 0..32 x10 7.14 -0. ft2/s (m2/s) Exponents based on mean particle diameter (D50) ranging from sand to fine gravel Coefficient based on mean particle diameter (note that cs1 must be adjusted for SI units) Mean flow depth.21 x10 0.30x10 1.59 x10 9.48 (for English units) for different gradation coefficients and sizes of bed material.49) where: Dn = Size of the bed material for which n percent of a sediment sample is finer.73 cs3 -5 Gr = 3 cs1 1. Note that if sediment transport in SI units is desired.02 3.42 x10 7.0 2.86 3.32 x10 6.0 4.1 provides the coefficient and exponents for Equation 4.0 5.87 -6 -6 -6 -6 6.1.196 -0.33 cs2 3.89 3.* 0. 50.

A description of the development and validation of this equation is presented in Appendix B.040 m/m Unit Discharge.50) where: qt V y S a. The range of parameters utilized to develop this equation is shown in Table 4.96 m2/s Particle Size. (1999) modified a Posada (1995) equation using nonlinear optimization and the field data for different sizes of riverbed sediment.12 (SI) and Section 4. and d = = = = = Sediment transport rate. 4.48 Developed by Simons et al.29 .13 (English). The resulting equation is: q t = aV b y c S d (4. If conditions are within the ranges outlined in Table 4. Equation 4. (1981).062 mm As the depth increases for a given velocity.2. Julien (1995)].2.6. Table 4. Range of Parameters Equation 4.48 was developed for channels with steep slopes for sand and fine gravel beds experiencing critical and super-critical flows [Simons. care should be taken to assure that the range of parameters being used is not out of range with those used to develop the equations (Table 4.2). If this is not true. metric tons/m/day (tons/ft/day) Mean flow velocity. There are several other checks that should be made in order to ensure the equations are applicable to a given problem.1. The equations are not applicable when armoring occurs.005 – 0. q 3. the intensity of the turbulent transfer properties decreases for these sizes. m/s (ft/s) Mean flow depth. they do not apply to conditions when the bed material is cohesive.2 the regression equations should provide results within ten percent of the values computed using the MeyerPeter and Müller bedload and Einstein suspended load equations for the steep conditions in Table 4.98 – 7.3 Expanded Power Function Relationship Kodoatie et al. c.05 – 60. et al. The equations are based on the assumption that all the sediment sizes present can be moved by the flow. b. The sizes with no dependence (Cs2 = 0) on depth in their transport rate fall between these two extremes. The result is an inverse dependence of transport rate on depth for these larger sizes.48 would overpredict transport rates in a cohesive channel.2. armoring will take place. m (ft) Energy slope Regression coefficients 4. An example of the application of the basic power function relationship is given in Section 4. Parameter Value Range SI Units Froude Number 1–4 -Velocity 1.92 m/s Bed Slope 0. When applying the equations resulting from Table 4. Since the equations were developed for sand-bed channels. D50 > 0.Equation 4. The increase in area available for suspended sediment with the increased depth does not totally counterbalance the reduced turbulent transfer characteristics.

is: log C t = 5. is: ωD 50 V log C t = 6. kinematic viscosity and sediment size.6 3.7 YANG'S EQUATIONS The Yang sand and gravel total load equations are presented because of their frequent application and wide acceptance. a* b Silt-bed rivers 281. The total load equations can be used to compute sediment transport by size fraction by using the geometric mean of the size class and weighting the computed concentrations by the class interval.4 3. depending on size of bed material.182 0.A summary of coefficient and exponents (for SI units) is presented in Table 4.3. Table 4.30 .3048(1+b+c) for input and results in English units.406 0.300 Gravel-bed rivers 431. which should be used for median sizes less than 2. 784 − 0 .000 d 0 0.0 and 10.8 1.622 Very fine to fine-bed rivers 2.1 x 0.409 log − 0.829. Yang (1996) related total load to excess unit stream power.4 2.681 − 0. Note that the values of "a" must be multiplied by a factor of 1. which should be limited to median sizes between 2. The sand equation.457 log * ν ω ωD 50 V ö æ VS Vcr S ö æ +ç .816 log * ν ω ωD 50 V* ö æ æ VS Vcr S ö ÷ +ç 2 .3048 (a) c 0.52) ç ÷ è ω ν ωø ω ø è 4.468 1. − 0. 305 log − 0 .314 log * ÷ logç − ÷ ç 1799 ν ω÷ ω ø è ø è ω ωD 50 (4.1 x 0. Coefficient and Exponents for Equation 4. The regression equations are developed based on dimensionless combinations of unit stream power.3.123.0 mm.000 (1+b+c) *a (English Units) = 1.435 − 0.0 mm. The Yang equations are also readily adaptable to computer solutions. expressed as the product of velocity and slope.412 0. Yang also developed critical velocity formulas for use with his equations. fall velocity.286 log V − 0. Separate equations were developed for sand and gravel bed material and solved for sediment concentration in ppm by weight.646 Medium to very coarse sand-bed rivers 2.51) and the gravel equation.50.613 2. critical unit stream power.000 An example of the application of the expanded power function relationship is given in Section 4. shear velocity. 282 log logç − ÷ (4.13 (English) 4.884.633 log − 4.12 (SI) and 4.

the dimensionless critical velocity is given by: Vcr 2. or mass or weight rates (metric-tons/day.53) and Vcr V* D 50 = 2.56) A common unit for sediment concentration is milligrams per liter (Cmg/l). m/s (ft/s) Velocity.where: Ct ω ν V* V Vcr S = = = = = = = Sediment concentration in parts per million by weight Fall velocity of the sediment.54) An example of calculating total bed-material discharge using Equation 4. Four ways of expressing concentration are: Concentration by volume: Cv = sediment volume/total volume Concentration by weight: Cw = sediment weight/total weight = CvSg / [1+(Sg-1)Cv] Concentration as parts per millions by weight: Cppm = 10 Cw Concentration as mg/l: Cmg/l = 106CvSg 6 (4.57) (4. 4. ft3/s). These include rate of transport and concentration.06 ν (4.12 (SI) and Section 4. Concentrations are often expressed as parts per million by multiplying the concentration by 106. m/s (ft/s) Critical Velocity. m2/s (ft2/s) Shear velocity ( gRS ) . m/s (ft/s) Energy slope In the above equations.05 for ≥ 70 ω ν (4.8 CONVERSION FACTORS There are numerous ways of expressing sediment transport. Qs (metric-tons/day) = Qs (m3/s) x γs / g x 86400 / 1000 Qs (tons/day) = Qs (ft3/s) x γs x 86400 / 2000 = Qs (ft3/s) x γs x 43.2 (4. The concentration relates volume or weight of sediment to the total volume or weight of the sediment-water mixture and is converted to a rate of sediment transport by including the water discharge. < * 50 < 70 = + 0. The following equations can be used to convert from a volumetric rate to a mass transport.60) 4.51 is given in Section 4.5 VD . m/s (ft/s) Kinematic viscosity.58) (4. which is the ratio of the mass of sediment to the total volume of water and sediment.59) (4.66 for 12 VD ω ν log * 50 − 0. tons/day).55) (4.13 (English). The rate of sediment transport can be expressed as a volumetric rate (m3/s.31 .

an equation should be selected based on the particular river bed material and flow characteristics. The review tested the equations with a large compilation of field data that encompasses a wide range of bed material. however. 4. and a wide range of river sizes.000 there is less than a 7 percent difference between Cppm and Cmg/l. There are several sources for obtaining recommendations on the applicability of sediment transport equations. The equations presented in this section have broad application. The difference is less than 1 percent for concentrations less than 10. an approximate rate of sediment transport can be estimated by Qs = Cppm x 10-6 x Qwater / Sg where Qs and Qwater are in m3/s or ft3/s. Another source for recommendations on suitability of various transport functions is the SAM Hydraulic Design Package for Channels (Thomas et al. ∆t. The BRI-STARS manual (Molinas 2000) presents a comparison of several sediment transport equations with a variety of measured data. from silts to gravel. If more than one equation appears suitable. Appendix B includes a review of ten sediment transport equations.62) When sediment is eroded or deposited. and should also be considered. For low concentrations. The volume of eroded or deposited material for a time interval. Army Corps of Engineers) refers to Vanoni (1975) as a guide to the applicability of sediment transport equations. 2000). or porosity. At these low concentrations. When possible.32 . is V = Qs x ∆t / (1 .S.(Note that Cmg/l equals the ratio of the mass of sediment in milligrams to the volume of the water-sediment mixture in liters. This is because the weight of the sediment is small compared to the weight of the water. Table 4.η) (4. the results of the sediment transport calculations should be compared with measured sediment transport. Sg is the specific gravity of the sediment (γs/γ).) In these equations. from 1 m width to over 1000 meters in width. they are often considered equivalent. is the ratio of voids to the total volume and often ranges from 35 to 45 percent. This is most easily done using sediment transport programs such as BRI-STARS.9 APPLICATION OF SELECTED SEDIMENT TRANSPORT EQUATIONS No single sediment transport equation can encompass all alluvial channel conditions.63) where Qs is the rate excess (or deficit) sediment transport and η is the porosity of the bed material. For sediment concentrations (Cppm) less than 100. the volumetric transport rate must include the void space between the sediment particles. HEC-6 and SAM. The HEC-6 manual (U.4 includes a list of sediment transport relationships and the application of each equation. Therefore. The void space. 4. one should apply the equations and determine which one provides the most reasonable results.000 ppm or mg/l. The exact conversion is: Q s = C ppm x 10 −6 x Q water /(S g (1 − C ppm x 10 −6 )) (4. other equations can.

Summary of Applicability of Selected Sediment Transport Formulas. Toffaleti (1969). 3 Karim X X 4 Karim and Kennedy X X 4 Kodoatie et al. Brownlie (1981). Toffaleti (1966) and Yang (1996). Karim and Kennedy (1981). Einstein (1950). Colby (1964). The power relationship can include depth. Schoklitsch (1937). modified Laursen equations (1958). Brownlie (1981). and Yang (1996). The U. and a user specified power relationship that is a power function of the product of depth and slope. 1975). 3. Molinas and Wu X 1 1. and Yang 1984) and a user specified power relationship. sediment size and discharge. 2. (Power) X (fine) X 4 Toffaleti X 2. Parker (1990). Meyer-Peter and Müller (1948). Einstein (1950). 3. 3. (Power) X X X 4 Laursen X (fine) X 2. several modifications to Laursen (1958). Schoklitsch (1937). Ackers and White (1973). 3. Duboys (in Vanoni. Ackers and White 1973. Formula Gravel Sand Very Fine Sand See and Silt Footnotes Ackers-White X 1. 3. 3 Molinas and Wu X 1 Parker X 3 Schoklitsch X X (coarse) 2. 2. The HEC-6 model includes the Toffaleti (1966). Colby (1964). 3 Shen and Hung X 4 Simons et al. and Kodoatie (1999). 3 Duboys X (fine) X 2 Einstein X X X 3. Appendix B of this manual includes an evaluation by Kodoatie (1999) of ten equations. 4. 2000) includes Ackers and White (1973). Meyer-Peter and Müller (1948). 4 Bagnold X X 4 Brownlie X 3. 4 Engelund-Hansen X 1. two gravel equations (Meyer-Peter and Müller 1948. 4 Meyer-Peter and Müller X 1. Laursen (1958).33 . 4. 4 Colby X 2. including Ackers and White (1973). In HEC-6. and Molinas and Wu 1996). velocity. Army Corps of Engineers SAM package (Thomas et al. Molinas and Wu 1996). slope. 2. Bagnold (1966). Yang (1973). The BRI-STARS model includes one fine sand equation (Yang.4. Karim (1998).Table 4.S. Posada (1995). the Colby (1964) correction factor for wash load concentration can be applied to most of the transport functions. Yang 1973. 4 Yang. 4 Yang X X 1. (1981). four sand equations (Engelund and Hanson 1972. Shen and Hung (1972). Appendix B also discusses the power function relationships by Simons et al. 2. Engelund-Hansen (1972).

For important bridges. BRI-STARS (Molinas 1990. The Colby procedure (Section 4.1) has been widely used to estimate bed load (contact load) in coarser bed systems. the results should be compared to assess the range of possible outcomes. 4.4) provides a reliable quick estimate of the bed material discharge in sand bed streams and the MeyerPeter and Müller bed load equation (Section 4.4. sources of sediment and modes of sediment movement.7) are widely accepted and adaptable to both hand calculation and computer solutions. determining the quantity and gradation of sediment being transported can be used to: • • • • • • Check on the previous determination of long-term aggradation or degradation and contraction scour. Typical applications of sediment transport relationships include equilibrium slope and sediment continuity analyses to determine long-term trends in bed elevation change (aggradation/degradation). Finally.10 SUMMARY Computing the bed material sediment transport capacity for a river should be grounded on a firm understanding of the river channel characteristics. 2001). 2001). If aggradation or if more refined degradation estimates are required.1 Step 1: Determine if Sediment Transport Computations are Necessary Determine qualitatively if the sediment problem at the bridge is aggradation or degradation. For a specific river. 4. then calculating the quantity and gradation of the sediment being transported may be necessary. previous scour calculations may be sufficient or a sediment transport analysis using equilibrium slope or sediment continuity could provide an initial estimate of long-term degradation (see HEC-20. Determine if there will be environmental problems upstream or downstream of the bridge from sediment transport conditions at the bridge crossing. If degradation. Estimate the cost of methods to solve the aggradation or degradation problem. objectives.11. Army Corps of Engineers 1993).6) provide a practical method for quick sediment transport calculations for sites within the range of conditions for which they were developed.5. and resources available for a specific project. Equations that are well suited for the particular river conditions should be used and. Lagasse et al. many of the sediment transport relations of Table 4. it is recommended that the results be compared with actual measurements.11 SEDIMENT TRANSPORTATION ANALYSIS PROCEDURE 4. 2000).4 are frequently used in sediment transport investigations and several are represented as options in the code of sediment transport models such as HEC-6 (U. The Yang sand and gravel equations (Section 4. The power function relationships (Section 4. Determine the type of countermeasures to solve a sedimentation problem. These applications are illustrated in HEC-20 (Lagasse et al.S. if more than one are well suited. The next section outlines a general sediment transport analysis procedure.5. Design debris basins for aggradation problems. Selection of an equation for use depends on the data available and scope. Design a check dam for degradation problems. 2000).34 . and SAM (Thomas et al.

Refining estimates of long-term aggradation or degradation will require multi-year cumulative quantities. computer models.35 .2 Step 2: Determine Scope of Sediment Transport Analysis Determine if the need for information on the quantity of sediment being transported is for the quantity transported by the occasional high or peak flows or for the amount transported annually. Field measurements used in conjunction with the modified Einstein method (Colby and Hubbell 1962. 4. Basic Methods for Sediment Transport Calculations The methods and equations given in this chapter will normally be sufficient for most determinations of the bed-material transport for a given discharge or to develop a bedmaterial discharge rating curve to be used with a flow duration curve to determine the annual sediment discharge. only sediment transported by high or peak flows may be needed.e. Simons and Senturk 1992) provide the most accurate method of determining the total sediment discharge of a stream. The design of a check dam or debris basin may require annual or multi-year cumulative quantities of sediment transport. Army Corps of Engineers are specialists in this type of field measurement.S. The USGS and the U.Additional issues that need to be addressed include: • • Is the cost of a sediment transport analysis commensurate with the cost of the bridge and the accuracy of the analysis? Are there sediment transport data available from the USGS or other Federal or state Agencies for this stream or similar streams? 4. i. • • • To check previous scour calculations. determine the sediment transport design parameters. and basic methods given in this chapter.11. Field Measurement Field measurement to determine sediment transport is a specialized activity and generally should be contracted for.11. The basic equations and methods given in this chapter and recommended for practical applications are: • • Meyer-Peter and Müller (1948) bed load equation (also USBR 1960) Colby’s (1964) curves and method 4.. Only for the most important and costly bridges would it be necessary to use field measurements and/or computer models to determine bed-material transport.3 Step 3: Determine the Type of Analysis The types of analyses available to determine the quantity and gradation of sediment transport are field measurements. Vanoni 1977.

4).36 . Army Corps of Engineers 1993) These computer models have the ability for user-selected sediment transport relations (including most of those shown in Table 4.• • • Simons et al. Colby's Curves and Method Colby’s method is only for sand size bed material (D50 from 0. The gravel equation should be limited to median sizes between 2.2). It is very good method for rapid determination of total bed-material discharge. Kodoatie et al. (1999) expanded power function method Yang (1996) sand and gravel equations A summary of each method is given below to aid in selecting which method to use. Two of the better known computer models are: • • BRI-STARS (Molinas 2000) HEC-6 (U. (1999) Power Function Relationship The Kodoatie method (modified Posada (1995)) can be used for streams with bed material size ranging from silt to gravel. (1981) basic power function method Kodoatie et al. Simons et al.1 – see Section 4.0 mm. medium to coarse sand and gravel bed (Table 4. Computer Models The use of computer models is beyond the scope of this manual.S.0 and 10.6 to 2. The difficulty in the method is the need to interpolate between sand sizes and gradation coefficient. (1981) Power Function Relationship The Simons et al. It is. best utilized for coarser bed material streams (D50 coarser than 2. It provides a coefficient and exponents based on classifying streams as silt. The relationship takes into consideration the size distribution of the bed material (Table 4.0 mm) that normally exhibit critical or supercritical flow.6) and power relationships. therefore. The models can be used to determine the total sediment discharge for a given flow or to 4. Meyer-Peter and Müller Bed Load Equation The Meyer-Peter and Müller bed load equation only calculates the bed material moving in contact with the bed (contact load). The method can be used to determine the total bedmaterial transport by size fraction. fine sand.1 to 5. It can be used to check or compare with the other methods.00 mm). power relationship can be used for steep streams with sand and fine gravel-size bed material (D50 from 0. It does not give total bed-material discharge unless there is no bed-material in suspension. The method is useful for rapid calculations of bed-material discharge.00 mm). but programs are available. The equations can be applied by size fraction or to the median size. Yang (1996) Sand and Gravel Equations The Yang equations can be used for sand and gravel total load estimation.

whereas HEC-6 is better suited for long-term simulations and should be used with caution for single event simulations.000210 Temp = 22.0338 m / s υ = 1. The SAM package uses normal depth hydraulics and does not perform sediment routing.12: éy − y c a ù ω =ê o ú and z = ca ë y yo − aû βκV* Z 4. 4.1 Introduction The following example problems illustrate the application of concepts and equations for sediment transport.12 SOLVED PROBLEMS FOR SEDIMENT TRANSPORT (SI) 4. BRI-STARS can be used for single event or long-term sediment transport simulations. It is intended for basic sediment transport analysis and for developing sediment rating curves as part of a sediment yield analysis. Given: y o = 4. some of the examples will only be given in English units.5°C (72. This rating curve is used in combination with a flow duration curve or table to determine the average annual bed-material discharge. and convert the results back to SI units.5°F) D 50 = 0. Readers who use metric units (SI) are encouraged when encountering similar problems to convert the input variables to English units.37 .30 m V = 1. calculate the vertical suspended sediment concentration profile for the D50 sediment size in mg/l.S.12. Because several of the equations are commonly given in English units (for example Einstein’s equation) and the methodology is very complex.000 mg / l From Equation 4.determine the average annual sediment discharge.27 m / s S = 0. The U. Army Corps of Engineers SAM package (Thomas et al. To determine the average annual total bed-material discharge the computer model is used to determine the bed-material discharge for a range of discharges to develop a sediment discharge rating curve or table. solve the problem.000255 m ω (D 50 ) = 0.00 x 10 −6 m 2 / s c a = 775.12.2 Problem 1 Suspended Sediment Concentration Profile From data observed on the Missouri River. 2000) includes sediment transport calculations using a wide variety of equations. 4.

12.8 mm 1.0941) Calculate the concentration versus distance profile given in Table 4.0338 = = 0.5 n = 0.38 .4) (0. Concentration vs. Given: A rectangular channel and the following data: D50 D90 G n nw yo Sf Dm W = = = = = = = = = 1.44 181 3.19) will be used.5: Table 4.41 0.0941 m/s ω 0.5. Bureau of Reclamation formulation of the equation (Equation 4.230 0.06 2.05 104 3.160 1.091 7.04 1.99 m 0.0) (0.01 mm 60.0005 m/m 2. Elevation Above the Bed. y (m) C (mg/l) 0.3 Problem 2 Using the Meyer-Peter and Müller Equation Calculate the Bed Sediment Discharge (Bed Load) In this example problem the U.Assume: β = 1.66 48 4.S.9 mm 2.898 βκV* (1.0 κ = 0.27 3 4.305 2.81) (4.00051 m Calculate: V* z= = (g yoS)1/2 = [(9.370 0.4 a = 2D50 = 0.61 1.00021)]1/2 = 0.96 m 4.22 531 2.30) (0.

65 metric-tons/day/m Qs = 5.06 ö 1+ ÷ ç 60.04 ) í1 + 1− ç ÷ ÷ý ÷ ç 60.19: é æ Qb qB = 4.5 = 0.84 é qB = 4.84) ê ë æ 2. solve the problem.038 ÷ è ø 3/2 ù (2.4.627 D m ú ú û 3/2 From Equation 4.12.Use Equation 4.5. and convert the results back to SI units.0005 ) − 0.627 (2.12.01)ú ú û 3/2 qB = 5.5 Problem 4 Calculation of Total Bed-Material Discharge Using Colby’s Method The application of Colby method to calculate total bed-material discharge is given in English units only. 4.038 ø 1.4 Problem 3 Application of the Einstein Method to Calculate Total Bed-Material Discharge The application of the Einstein method to calculate total bed material discharge is given in English units only.13. See Example Problem 4. See Example Problem 4.04 ø øï ï î þ 2/3 = 0.21: 3/2 ü ì 5.780 ê10. Organizations using SI units are encouraged to convert the input variables to English units. 8 1 / 6 ö ç ÷ ç 0.96 = 344 metric-tons/day 4.06 ö ö ï ï ç n b = (.13.96 è 0. The Colby method is easy to use.23: Qb = Q 1 5.846 ç ç Q ê è ë D 90 öæ ÷ ÷ç ç ø è nb 1/ 6 ö ÷ ÷ ø 3/2 ù y o S f − 0.780 ê10. The example problem uses the hydraulic and sediment data given by Einstein for the purpose of comparison of the two methods. 4.846 (0.038 From Equation 4.39 .98 æ 0.99 ) (.65 x 60. The example is taken from Einstein’s 1950 publication.98 æ æ .96 è è .

83 m3/s.31 mm and size distribution factor G =1.64 Cs1 = 1.613 qs = 2123.63 x 10-5 (English) Cs2 = 0.468 (0.40 .65 x 9810/9.30 (1.1 Cs1 = 1. Sediment properties of D50 = 0.3048 ( 2−Cs 2 −Cs 3 ) Cross-interpolate in Table 4.402 /1000 = 92. Depth y = 1. Depth y = 1.438 m/s. Width W = 60.64) = 1.81 x 3600 x 24 x 0. Compare the result with the expanded power function application in Problem 6.3 From Table 4. (1999) Equation 4.63 x 10-5 x 0.96 m.32.12. Q = 271.32. Simons et al.83 m3/s.7 Problem 6 Calculation of Total Bed-Material Discharge Using the Expanded Power Function Relationship Determine the bed-material discharge for the 100-yr discharge for a stream with the data given in Problem 5.468.6 Problem 5 Calculation of Total Bed-Material Discharge Using the Basic Power Function Relationship Determine the bed-material discharge for the 100-year discharge for a stream with the following data.3 a = 2123.45-3.613 = 518 metric tons/m/day 4. c = 0.96 m.000521. (1981) Equation 4.829 m.000521)0. Velocity V = 2.96 x 0.64 = 0.0066 m2/sec Qs = 60. The data are repeated below.402 m3/sec Using a specific gravity of 2.3048(2-0. S = .4 (2.438 m/s.3048-2.829)0. Velocity V = 2.4. Note that these data are not in the Fr range of Table 4.438)3. Sediment properties of D50 = 0.50 is qs = a Vb yc Sd The sand size places this in a medium sand bed stream in Table 4.12.2.65 Qs = 2.63 x 10-5 x 0.000 metric-tons/day 4.000195 (1. d = 0.1 and cross-interpolate to obtain the values of Cs2 and Cs3.0066 = 0.31 mm and size distribution factor G =1.438)3.45 Cs3 = 3.45 (2. Width W = 60.829)0. For Cs1 in SI units use the following Equation Cs1 (SI) = Cs1 (English units) x 0. Kodoatie et al.4.48 is q s = c s1 y Cs 2 V Cs 3 Use Table 4.829 m.30.000195 (SI) qs = 0.09 =0. b = 3. Q = 271.

5 = + 0.06 ν and Vcr VD = 2.12.00022 m/m ν = 1.32 mm S = 0.00059 ω VS = 0.00472 ω W = 103.95 m D50 = 0.922 x 0.043 = 0.69 x 0.5 − 0.922 m/sec Calculate: τo = 9800 x 2.800 tons/day 4.043 m/sec 4. < * 50 < 70 = + 0.409 log æ è ωD 50 ν − 0.8 Problem 7 Calculate Total Bed-Material Discharge Using Yang's Sand Equation Yang's sand equation (Equation 4.98 N/m2 V* = (4.66 for 12 V D ω ν log * 50 − 0.580 metric-tons/day = 31. − 0.31 x 0.314 log V* ö æ VS ÷ logç è ω ωø − Vcr S ö ω ÷ ø Vcr 2.00032 / 1.05 for * 50 ≥ 70 ω ν Compute the bed-material transport in metric-tons/day.62/2000 = 34.580 metric-tons/day Note Qs = 31.69 ω log 19.435 − 0.5 VD .286 log ωD 50 ν − 0.16 x 10-6 m2/s ω = 0.580 x 2204.00022 / 0.0706 x 0.39 m3/sec y = 2.96 = 31.00022 = 4.0706 m/s V* D 50 = 0.16 x 10 − 6 = 19.5 ν Vcr 2 .00022 = 0.06 Vcr S = 2.66 = 2.457 log V* ω + ç 1799 .98/ 1000)0.51) is as follows log C t = 5.5 = 0.31 m V = 0.Qs = 518 x 60.41 . Given: Q = 221.

86 − 0.000838 ft ) ω (D 50 ) = 0.V* = 0.314 log 1. 4.3 x 89 x 3600 x 24/106 = 1700 metric-tons/day 4.017 x 10 −5 ft 2 / s c a = 775.00059 ) = 1.000210 Temp = 72.5°F (22.3 lb / ft 3 ) From Equation 4.286 log 11.13.12: éy − y c a ù ω =ê o ú and z = ca ë y yo − aû βκV* Z 4.13 SOLVED PROBLEMS FOR SEDIMENT TRANSPORT (ENGLISH) 4.13.86 ν log C t = 5.64 ω ω D 50 = 0.00472 − 0.435 − 0.0706 / 0.799 − 0.164 ft / s S = 0.1 Introduction The following example problems illustrate the application of concepts and equations for sediment transport.043 = 1.86 − 0.1109 ft / s υ = 1.64 + (1.64 ) log ( 0.42 .00032 / 1.5°C) D 50 = 0.16 x 10 −6 = 11.409 log 11.2 Problem 1 Suspended Sediment Concentration Profile From data observed on the Missouri River.95 = 89 ppm by weight Qs = 221.255 mm (0.000 mg / l ( 48.95 C t = 10 1. calculate the vertical suspended sediment concentration profile for the D50 sediment size in mg/l. Given: y o = 14.10 ft V = 4.043 x 0.457 log 1.

0017 ft Calculate: V* = (g yoS)1/2 = [(32. Elevation Above the Bed.6: Table 4.4 a = 2D50 = 0.8 ft 0.1109 = = 0. Bureau of Reclamation formulation of the equation (Equation 4.13.06 9. y (ft) C (mg/l) 0.S.19) will be used.5 n = 0.04 1.1) (0.0) (0.300 1 2.180 4 538 8 183 10 105 12 49 14 3 4.8 mm 1.43 .6.898 βκV* (1.0 κ = 0.0005 (ft/ft) 2.2) (14.Assume: β = 1. Concentration vs. Given: A rectangular channel and the following data: D50 D90 G n nw yo Sf Dm W = = = = = = = = = 1.309 ) Calculate the concentration versus distance profile given in Table 4.309 ft/s z= ω 0.3 Problem 2 Using the Meyer-Peter and Müller Equation Calculate the Bed Sediment Discharge (Bed Load) In this example problem the U.41 0.360 2 1.4)(0.00021)]1/2 = 0.3 7.01 mm 200 ft 4.9 mm 2.

4.606 ê3.20.01)ú qB = 1.21: 3/2 ü ì æ . 8 1 / 6 ö ÷ (9. hydraulic radius and wetted perimeter versus stage for the representative cross section are given in Figure 4. The relations of the cross-sectional area. The sediment transport calculations will be made for individual size fractions with selected representative grain sizes equal to the geometric mean grain diameter of each fraction.84 ) ç ç 0.038 ÷ ú ê è ø û ë 3/2 qB = 1. The reader can refer to the original example for the construction of the representative cross section and the consideration of bank friction. the wetted perimeter is assumed to equal the surface width.7.4 Problem 3 Application of the Einstein Method to Calculate Total Bed-Material Discharge A test reach. For simplicity.147 mm. For this wide and shallow channel. The characteristics of the channel cross-section follow.84 3/2 ù é æ 2.Use Equation 4.038 ø 1.8 ) (.06 x 10-5 ft2/sec. His numerical example is reproduced here.8% of the bed material falls between 0. The kinematic water viscosity ν is 1.306 ç ç Q ê è ë D 90 öæ ÷ ÷ç ç ø è nb 1/ 6 ö ÷ ÷ ø 3/2 ù y o S f − 0. The channel slope was determined to be S = 0. 95.606 ê3.89 x 200 = 378 tons / day 4.627 (2. the effects due to bank friction are neglected.23: Qb = Q 1 19.6 æ 0.306 (0.589 and 0.89 ton / day / ft Q S = 1.19: é æ Qb qB = 1.06 ö 1+ ÷ ç 200 è 0.04 ) í1 + 1− ç ÷ ÷ý ÷ 200 ç ï è è .04 ø øï î þ 2/3 = 0. representative of the Big Sand Creek near Greenwood. Note that of these composite samples.627 D m ú ú û 3/2 From Equation 4.06 ö ö ï ï 19. which is divided into four fractions.038 From Equation 4. and the grain size distribution is presented in Figure 4.00105.13. and the specific gravity of the sediment is 2.65. Mississippi was used by Einstein (1950) as an illustrative example for applying his bed-load function.0005 ) − 0.5 = 0.44 .19. The averaged values of the four bed-material samples are given in Table 4.6 æ ç n b = (.

00162 17.195 0.148 0. The procedure and results are given in Table 4. their meanings and calculations are explained after the table.8.70 0.147 > D --1. as given in Table 4.295 0. In the calculations.7 were used to evaluate the resistance to flow.147 0.Table 4.7.75 0.00057 5.9.42 and Figure 4.248 0.175 0.8 --Hydraulic Calculations The important hydraulic parameters were calculated.208 > D > 0.417 > D > 0.20 0.495 0.417 0. see the notes section for explanation of symbols. Any similar relation discussed in Chapter 2 can be utilized for the same purposes. Grain Size Average Grain Size Settling Velocity Distribution.45 .2 3.00081 32. Description of the average cross section. 4.8 5.064 0.351 0. column by column.208 0.589 > D > 0.8 1.0 2. Equation 4.100 0. Bed-Material Discharge Calculations The bed-material transport is calculated for each representative grain size of the bed material at each given flow depth.295 > D > 0.589 --2.70 0. Bed Material Information for Sample Problem.19. The table headings. mm mm ft % mm/sec fps D > 0.4 --0. Figure 4.00115 40.

Figure 4. Grain size distribution of bed material.20.46 . 4.

27 22 10.76 2.400 16 103 136 170 194 234 289 398 17 410 1. from Equation 4.0 Rb V′* 2 0. R′b 1 0.1 153.98 1.3 154.14 5.19 for the given stage (16) Pb ft.820 5.26 R′b /∆ fps (average flow velocity) (8) Ψ′ = (ρs .77∆ for ∆/δ′ > 1. Figure 4.36 1.0 51.00069 0.412 0.50 0.04 (1) R′b ft (bed hydraulic radius due to grain roughness).14 0.7 (6) ∆ = ks/X ft (apparent roughness diameter) (7) V = V′* 5. Application of the Einstein Procedure (after Einstein 1950).21 1.00039 0.30 0.44 " (9) V/V * = f (Ψ′) given in Figure 4. given by Equation 4.91 0.70 11.91 0.10 11.19 0.2 y /∆).259 0.40 11. given in Figure 4.0 87.368 0.00072 0.36.07 6.03 (β/βx) 2 " Rb 11 0.318 0.36 1.50 3.0 240.84 4.27 1.13 0.44 2.85 1. ft .27 1.91 0.95 3.90 12.8 (20) β x = log (10.184 0. ft.07 0.54 0.6 X /∆).8 (19) Y = f (ks/δ′) (pressure correction term).16 0.5 1. given in Figure 4.303 log (30.8 27.76 0. (cross-sectional area).86 0.63 0.84 0. ft (roughness diameter) (5) X = f (ks/δ) (correction factor in the logarithmic velocity distribution).00033 0.03 Notes: 13 1. from description of cross section. a stage-discharge relationship can be plotted by relating the computed Q to the stage (18) X ft (characteristic distance).32 4.50 3.00082 19 0.065 2.30 4.00078 0.55 0.50 0.9 156.72 2.00095 0.26 A X 5 1.129 0.14 5. given by Equation 4.0 370.00067 0.ρ /ρ ) (D35 /R′bS) (intensity of shear on representative particles).00097 0.50 3.00075 0. with no additional friction from the banks. (bed wetted perimeter).30 0.03 PE " 12 1.03 14 150.00084 0.44 6.19 for R = Rb.00027 Stage ks/δ′ 4 1.00079 0.54 0.450 y = Rb δ′ 3 0.14 1.75 log (12.59 1.00091 0. values are assumed to cover the entire desired discharge range (2) V′* = gR b S fps (shear velocity due to grain roughness) ' (3) δ′ = 11. Rb represents the total hydraulic radius R (13) y ft (average flow depth).40 9.0 2.620 16.00093 0.00104 0.8.27 1. (logarithmic function) (21) β = log 10.5 15 140 240 425 640 970 1.54 20 1.91 0.07 6. from Figure 4.29 1.27 1.0 4.9 152. y ≈ Rb for wide shallow streams (14) Stage.39δ′ for ∆/δ′ < 1.92 11.63 8.56 0.97 11.68 0.92 4.37 0.49 0. from Figure 4.550 30.0 βx " V* 10 0. 2 (15) A.9 159.30 12.19 for the given stage (17) Q = AV cfs (flow discharge).76 2.0 6. (Einstein’s transport parameter).07 0.380 9.91 21 0.Table 4.465 2.25 Y V/V * 9 16.46 1.27 1.2 150. X = 0.00047 0.05 0.30 4.58 X Ψ′ 8 2.00030 0.0 3.00101 0.18 1.10 0.17 0.11 1. Hydraulic Calculations for Sample Problem 3.11 " fps (shear velocity due to form roughness) (10) V * (11) R"b ft (bed hydraulic radius due to form roughness) from V"* = gR b S " (12) Rb = R′b + R"b ft (bed hydraulic radius).6 (22) PE = 2.30 11.08 Pb ∆ 6 0.0 5.6ν / V′* ft (thickness of laminar sublayer) (4) ks = D65.00107 Q V 7 2.0 150.00132 0.80 and X = 1.200 18 0.75 0.47 .

10 1.25 1.44 1.50 39.1800 0.43 0.01 1.0 623.44 1.86 0.88 2.01710 0.350 6.000 324.58 5.2070 0.00 26.050.90 0.29 2.3800 0.45 5.400 57.020.18 1.70 8.50 25.03450 0.13000 3.07540 0.50 41.50 25.25 0.0 483.000 iBQB 10 119.49 0.54 1.37 1.35 0.5 1.80 0.0 0.0 4.23 4.25 0.00 1.31000 0.3490 0.70 0.17 1.20 28.54 1.500 59.00 ξ 6 1.39 0.5 29.36 1.0 2.88 1.09580 0.90 0.0 0.56 0.63 0.15 1.72 0.0 313.510.2530 0.85 0.85 4.78 0.0471 0.28 1.78 2.30 17.00081 0.700 157.80 15.97 0.00 516.230 0.25 0..000 153 1.90 0.560 0.60 19.90 0.50 39.00 Ψ* 7 2.44 1.720 2.00 8.50 36.75 6.68 0.01 2.540.92 0.63 0.810 1.44 0.000 800.60 0.26000 4.14 3.3640 0.04260 å iTQT 19 670 3.18 iTqT 17 0.0 3.0 4.000 iBqB 9 0.10 21.08 2.0 5.30 0.17 5.50 27.01 0.200 9.16 2.38 1.50 iTQT 18 168 561 2.05 1.99 0.68 1.50 46.25 0.0 3.0 4.450 0.75 6.60 2.08 1.2790 0.2420 0.000 14 587 9.000 335.00 1.0 1.000 70.00 1.51 0.0 4.82 1.00 1.69 1.19 4.058 å iBQB 11 400 1335 3.02460 0.090 0.950 0.92 0.20 26.0 6.50 31.10 1.3160 0.360.19 2.000 1.57 1.Table 4.20 12. Bed-Material Load Calculations for Sample Problem by Applying the Einstein (1950) Procedure.00 31.03870 0.05 2.79 0.0 6.39 0.0 12.0 845.520 1.03 1.0 0.00057 0.46 0.178 R′b 3 0.10 17.91000 2.04 1.0 330.00 0.42 0.4 126.350 37.29 2.940.17 1.18 0.00 22.33 1.00 74.94 1.928 30.0 3.48 .44 3.43 0.65 1.20000 2.050 5.10 11.6040 0.98 0.90 4.38 1.00 3.0 69.00 5.70 56.500 113.0 3.350.98 0.61 0.330 6.03800 0.00 2.35 2.660 20.32 0.496 10.0 4.000 D/ X 5 1.0 732.00 1.0 2.80 18.50 43.370 2.41 0.00 92.80000 ϕ* 8 1.33 0.08 1.23 1.42 1.210 0.131 0.40 2.00162 iB 2 0.0 6.00 31.0600 0.85 0.0 7.00 122.61700 1.05 2.00 312.49 0.0 1.980.61 0.771 6.09 0.27 0.27 0.27 0.0 2.10000 1.0 353.5000 0.800 248.402 0.37 1.69 0.000 1.050.50000 9.32 0.53 1.43 0.1150 0.100 129.27 2.01 0.0 1.0 3.00 722.84 1.83 1.5 1.65 1.1060 0.61 0.38 0.0 2.35 0.61 0.56 0.000 0.830 1. D 1 0.35 8.60 1.560.0 6.730.30 0.0 6.44 0.20 1.12 0.600 97.950.15 I1 14 0.20 3.70 PEI1+I2+1 16 1.34 0.28 0.49 0.00056 0.65 0.65 0.900.38 1.745 16.1390 0.500 49.9.800 335 1.31 0.460 12.5 1.400 19.000 51.71 2.16 12.800 15.00115 0.0 2.0 210.83 0.50000 0.73 11.74 1.80 3.00500 0.0 6.36 0.44000 5.33 3.00 146.74 2.54 1.4060 0.76 6.0 2.23 0.078 0.32 0.28 2.60000 0.117 0.21 1.80 23.0 Z 13 3.05 1.00000 32.78 9.0 2.35 1.240 0.00000 59.100 26.87 1.70000 5.7490 0.0 206.00 1.000 397.26 1.50 12.22700 0.170 12.12 1.0 0.10 10.142 103E 12 2.00 2.45 0.52 1.0 5.17 1.300 32.03 1.61 0.000 526.00 17.0561 0.30 0.440 8.42 3.00 63.35 Ψ′ 4 5.00000 30.20 0.0155 0.0 5.20 0.27 0.25000 1.333 27.780.48000 0.820.0 5.71 0.0 1.70 -I2 15 0.95000 15.60 6.0 5.02 1.19 0.00 1.40 183.44 4.16 1.00312 0.0267 0.5 1.120 2.40000 19.19 5.45 0.385 0.76 0.55 20.00 91.320 0.28100 0.85 0.85 5.0 3.73 0.42 0.08 2.50 1.55 1.68000 20.15 1.90 1.50 39.32 0.65000 10.020.04 1.22 0.63 0.530 3.

31 (17) iTqT is the bed material load per unit width of stream for a size fraction lb/sec-ft = iBqB (PEI1+I2+1) given by Equation 4.2WiBqB.7 (3) R′b is the bed hydraulic radius due to grain roughness. see Table 4.8 (8) ϕ * is the intensity of sediment transport for an individual grain (9) iBqB is the bed load discharge per unit width for a size fraction. (14) I1 is an integral given by Equation 4. given in Figure 4.9 (continued) Notes: (1) D is the representative grain size.ρ)/ρ) (D/(R′bS)) (5) D/ X for values of X. (13) Z is the exponent for concentration distribution = ω/(0.31 (18) iTQT is the bed material load for a size fraction for entire cross section. Column 18 (6) ξ = f (D65 / X) hiding factor.4 V′*) given in Equation 4. ft.7 and 4.41.8. given in Table 4. tons/day (12) E is the ratio of bed layer thickness to water depth = 2D/y.8 (4) Ψ′ is the intensity of shear on a particle = ((ρs . ft.Table 4.6 (7) Ψ* is the intensity of shear on individual grain size = ξY (β/βx)2 Ψ. given in Table 4.40 and read from Figure 4.9 (15) -I2 is an integral given by Equation 4. Values of ω and V* are given in Tables 4. tons/day = 43.2 WiTqT (19) å iTQT is the total bed material load for all size fractions.8 (11) å iBQB is the total bed load discharge for all size fractions for the entire cross section. For values of y.8. W = Pb given in Table 4.7 (2) iB is the fraction of bed material. see Table 4. ton/day = 43.8. given in Table 4.49 . given in Equation 4.39 and read from Figure 4. tons/day 4.10 (16) PEI1+I2+1 is the factor between bed load and total load given by Equation 4. values of Y and (β/βx)2 are given in Table 4.33 (10) iBQB is the bed load discharge for a size fraction for entire cross section.

142 85. directly measured values of velocity and depth are required.8 qn tons/day-ft (uncorrected sediment discharge per unit width of channel).93 4. However. The Colby curves (Figure 4.13 for the given V. fps (average velocity).76 2.8 established by hydraulic calculations.13) are limited to 10 ft/sec.10 and 4.90 k2 6 1. The mean velocity and the depth of flow may be obtained by hydraulic calculations.38 9. Cf) (correction factor for fine sediment concentration). In addition. the sample problem used to illustrate Einstein’s method is solved using the Colby method.410 4.0 220. Table 4. y. y.22 1.5 50. Two procedures are presented.000 ppm. given in Figure 4. the water temperature and the fine sediment concentration are assumed to equal 70°F and 10. taken from Table 4.11 since the corresponding velocities are greater than 10 ft/sec.11 gives the calculations for individual fractions using the bed material size distribution.10.92 qn 4 14. the calculations are summarized in tables.36 1.50 3. whereas Table 4.500 47.8 V.44 6.90 0. respectively.5 Problem 4 Calculation of Total Bed-Material Discharge Using Colby’s Method In applying the Colby (1964) method.63 8. ft (surface width). y 1 1.99 0.25 k3 7 0.91 0.23 1. Table 4.0 325. as in Problem 3.20 1.14 qT = [1+(k1k2 -1) k3]qn tons/day-ft (true bed material discharge per unit width of stream).91 0.14 Notes: (1) (2) (3) (4) (5) (6) (7) (8) (9) y.0 k1 5 0. For convenience.07 ft and 6.13. The required data are taken from Table 4.0 135. given by Equation 4.99 qT 8 16 56 150 243 365 QT 9 1. T) (correction factor for temperature).46 QT = WqT tons/day (bed material discharge for all size fractions for entire cross section) W 2 103 136 170 194 234 V 3 2. the required data are: the mean velocity.30 4.8 are not shown in Tables 4.14 k2 = f (y. taken from Table 4.03 ft from Table 4.21 1. Note that depths. to obtain best results from calculations. the median size or size distribution of the bed material. given in Figure 4. given in Figure 4. taken from Table 4.4. the depth.50 .616 25.92 0.14 k3 = f (D50) (correction factor for sediment size).8 W. ft (mean depth).99 0. For purposes of comparison.648 7. taken from Figure 4.10 gives the calculations using the mean diameter of bed material. of 5. and D50 k1 = f (y. the water temperature and the fine sediment (wash load) concentration.99 0.99 0. Bed Material Load Calculations for Sample Problem by Applying the Colby Method (Median Diameter).

tons/day-ft (corrected bed material discharge per unit width by assuming the bed is composed entirely of one sand of size D) (11) ibqt.180 11.0 51.62 0.20 1.62 0.30 4.44 6.91 0. (bed material discharge per unit width for a size fraction for entire cross section) (12) ibQt = WibqT.0 0.0 85.4 19.91 0. given in Table 4. D. taken from Table 4.91 0.93 4.92 2.570 7.00 1.92 2.22 1.92 0.63 8. given in Figure 4.23 1.38 9.0 37.14 (8) k2 = f (y.90 0. tons/day.0 27.25 Notes: k3 9 0.50 3.90 0.91 0.91 0.0 6. given in Figure 4.0 130.00 1. Bed Material Discharge Calculations for Sample Problem by Applying the Colby (Individual Size Fraction).0 53.050 3.780 45.0 20.51 .560 77.8.14 1.8.00 1.97 D.44 6. Column 16 V.0 V 5 2.90 0.97 0.90 0.50 3.900 659 2.8. Column 13 W.93 4.92 0.23 1.7 ib (fraction of bed material). D 1 0.36 1.Table 4.3 4. taken from Table 4.00 1.97 0. tons/day-ft (incorrect sediment discharge per unit width by assuming the bed is composed entirely of one sand of size D) taken from Figure 4.76 2.14 (9) k3 = f (D50) (correction factor for sediment size).44 6.36 1. taken from Table 4.90 k2 8 1.25 1.38 9.14 1.92 0. (bed material discharge for a size fraction for entire cross section) (13) åibQT.91 0.63 8. mm (representative grain size.2 0.500 29.21 1.36 1. taken from Table 4.62 0.92 0.00 0.92 0.90 0.25 1.7 21.90 0.248 32.0 50.14 qT 10 13 43 119 205 288 16 49 132 230 323 20 58 155 266 388 25 70 180 337 471 W 4 103 136 170 194 234 103 136 170 194 234 103 136 170 194 234 103 136 170 194 234 IbqT 11 2.97 0.14 1.22 1.0 93.50 3.0 124.44 6.92 1.30 4.22 1.92 2. and y (7) k1 = f (y.23 1.22 1.4 20.76 2.880 6.0 1. tons/day-ft.21 1.580 8.76 2.97 0.92 0.351 40. (top width).92 0.91 0.7 y. (bed material discharge for all size fractions for entire cross section) (1) (2) (3) (4) (5) (6) 4.93 4. (average flow depth).93 4.14 (10) qT = [1 + (k1K2 -1) k3]qn.T) (correction factor for temperature).0 6.76 2.495 102ib 2 17.20 1.11. ft.420 659 2.25 1.8 y 3 1.908 22.700 3. given in Figure 4.30 4.23 1.000 30.50 3.38 9.Cf) (correction factor for fine sediment concentration).320 qn 6 12 40 112 193 265 15 45 120 210 290 18 53 140 240 345 23 64 163 305 420 åibQT 13 1.010 18.62 0.63 8.90 0. Column 7 qn.91 0.92 0.62 0. tons/day.175 5.5 4.500 16.20 1.640 k1 0. ft.21 1.13 by interpolation on logarithmic paper for the given V.3 7.92 IbQT 12 237 1.36 1.21 1.000 155 558 1. fps (average velocity).1 10.710 6.38 9.720 9.20 1.63 8.8 7 0.

45 (8.The results of the bed-material discharge calculations for the sample problem using Einstein's (1950) and Colby's (1964) methods are shown in Figure 4.52 . Sediment properties of D50 = 0. Simons et al. The curves indicate that the sediment discharge increases rapidly with an increase in water discharge.63 x 10-5 Cs2 = 0.00)3.21. In general.64 qs = 1.0)0.45 Cs3 = 3.32.62 = 91. (1981) Equation 4.31 mm and size distribution factor G =1. Note that these data are not in the Fr range of Table 4. 4.2. Depth y = 6. Cross-interpolate in Table 4. Q = 9600 cfs.0 ft.65 x 62. Width W = 200 ft.21.1: Cs1 = 1. Figure 4.600 metric-tons/day 4.14 ft3/sec Using a specific gravity of 2. Comparison of the Einstein and Colby methods. the two methods compare relatively well.63 x 10-5 (6.14) /2000 = 101.000 tons/day Qs = 101.0707 = 14.65 Qs = (2.1 and cross-interpolate to obtain the values of Cs2 and Cs3.0707 ft2/sec Qs = 200 x 0.0 ft/s. Compare the result with the expanded power function application in Problem 6.64 = 0. Velocity V = 8.4 x 3600 x 24 x 14.48 is q s = c s1 y Cs 2 V Cs 3 Use Table 4.13.6 Problem 5 Calculation of Total Bed-Material Discharge Using the Basic Power Function Relationship Determine the bed-material discharge for the 100-year discharge for a stream with the following data.000 tons/day x 2000/2204.

286 log ωD 50 ν − 0.50 is qs = a Vb yc Sd The sand size places this in a medium sand bed stream in Table 4. Depth y = 6 ft. Kodoatie et al.00022 ν = 1.4) = 8.53 . d = 0.468 (0.7 Problem 6 Calculation of Total Bed-Material Discharge Using the Expanded Power Function Relationship Determine the bed-material discharge for the 100-year discharge for a stream with the data given in Problem 5.800 tons/day 4.435 − 0.3. (1999) Equation 4.78 x 200 = 34. From Table 4.06 ν and Vcr VD = 2. The data are repeated below.000521.3 a = 2123.13. Sediment properties of D50 = 0. Q = 9.30+0.468) (2123.314 log V* ö æ VS ÷ logç è ω ωø − Vcr S ö ω ÷ ø Vcr 2.600 cfs.5 VD = + 0.1 x 0. < * 50 < 70 VD ω ν log * 50 − 0.51) is as follows: log C t = 5.00105 ft S = 0.32.05 for * 50 ≥ 70 ω ν Compute the bed-material transport in tons per day.095 qs = 8. Width W = 200 ft.30 (6)0.21 x 10-5 ft2/s ω = 0. Velocity V = 8 ft/s. b = 3.000521)0.02 ft/sec W = 341 ft D50 = 0.4.820 ft3/sec y = 7.3048(1+3.4.13.66 for 12 .457 log V* ω + ç 1799 .1 x 0. Given: Q = 7.141 ft/sec 4.58 ft V = 3.31 mm and size distribution factor G =1.613 a (English units) = 1.409 log æ è ωD 50 ν − 0.3048(1+b+c) a = 1.78 tons/ft/day Qs = 173. S = 0. c = 0. − 0.468.095 (8)3.613 = 173.30.8 Problem 7 Calculate Total Bed-Material Discharge Using Yang's Sand Equation Yang's sand equation (Equation 4.

4 x 3600 x 24 /(2000 x 10 6 ) = 1900 tons / day 4.67 x 0.00022 / 0.2 ν log C t = 5.232 = = 1.54 .1 ν Vcr 2.141 x 0.435 − 0.64 + (1.232 ft/s V* D 50 = 0.96 = 91 ppm by weight Q s = 7820 x 91 x 62.457 log 1. 5 = + 0.64 ω 0.00471 ω V* 0.314 log 1.00105 / 1.5 = 0.4 x 7.2 − 0.66 = 2.104/1.409 log 12.2 − 0.141 = 0.00105 / 1.Calculate: τo = 62.06 Vcr S = 2.04 lb/ft2 V* = (0.00022 = 1.94)0.96 C t = 10 1.00022 = 0.00471 − 0.64) log (0.141 ω D 50 = 0.58 x 0.67 ω log 20.121 x 10 −5 = 20.799 − 0.00059 ω VS = 3.232 x 0.00059 ) = 1.21 x 10 −5 = 12.1 − 0.286 log 12.02 x 0.

and rivers that are essentially straight. control works. in fact it is one of the most youthful rivers. Rivers have provided a means of traveling inland and developing trade. Specific aspects of stream form and a simple geomorphic classification of streams are presented later in this chapter.2 FLUVIAL CYCLES AND PROCESSES Fundamental characteristics and processes governing the formation of river systems are discussed in this section. braided rivers are relatively steep and meandering rivers have more gentle slopes. Mature. braided rivers.1 INTRODUCTION Rivers and river systems have served man in many ways.CHAPTER 5 RIVER MORPHOLOGY AND RIVER RESPONSE 5. and old (Davis 1899. and urban properties during floods. the flooding by rivers and the deposition of sediment on the river valleys have been a means of revitalizing the river valleys to keep them productive. the lower Mississippi River would be designated as old. This has played a significant role in the development of all countries wherever rivers of significant size exist. Meandering rivers that are not subject to rapid movement. More recently. Rivers have different alignments and geometry. To a large degree. a floodplain. 20 (Lagasse et al. and pattern (Brice 1982).000 years ago. are reasonably predictable in behavior. particularly in the arid and semiarid parts of the world. Rivers are passage ways for navigation and are essential to agriculture. when. bridges. meandering rivers are generally unstable with eroding banks which may result in destruction of productive land. the slope and energy of the streams are just sufficient to transport the materials delivered to it. 5. and alluvial fans is described as well as the processes of headcutting and nickpoint migration. As a general concept this has validity with steep irregular young streams becoming mature with a narrow valley. Nevertheless. 5. As time passes. and Old Streams One of the early methods to classify rivers was by relative age as youthful. Hydraulic Engineering Circular (HEC) No. 2001) 5. the concept of landform and channel evolution through time is a valuable one. The morphology of floodplains. It developed on the alluvium deposited by drainage from the continental ice sheet perhaps 10. The concept of geomorphic threshold completes this section on fundamentals leading to a discussion of variability and change in large alluvial rivers. the valley widens and a fully meandering channel of low gradient develops. Unfortunately. buildings.2. see King and Schumm 1980). deltas. Bank protection works are often necessary to stabilize certain reaches of many rivers and to improve them for other aspects of flood control and navigation. mature. that is.1 . bridge approaches. A very general classification of rivers which considers their age is introduced. and a graded condition.1 Youthful. In general. rivers have been classified based upon type of sediment load (Schumm 1977). for this classification. There are meandering rivers. however.

however. As the streams transport sediment to areas of flatter slopes. are characterized by unstable channel geometries and rapid lateral movement. and are subject to modifications due to climatological changes and as a consequence of human development.2 . The upstream river bed is filled in and average flood elevations are increased. 5. the main stream can be forced to make drastic changes at the time of major floods by the stream's tributaries. debris flows. Streams move laterally pushing the highlands back. the process of channel rejuvenation refers to an increase in erosional activities in mature or old channels caused by lowering base level elevation. the material is deposited forming deltas. and wet or fluvial fans formed by perennial stream flow. The second situation occurs when sediment material is moved through a fan-head trench and deposition occurs at the toe of the fan. They occur whenever there is a change from a steep to a flat gradient. and in particular to bodies of water where the velocity and turbulence are too small to sustain the transport of the material. area. are influenced by movements of the earth's crust. Consequently. The steep channel tends to drop part of its sediment load in the main channel building out into the main stream. including the long-term cycle of erosion and evolution of incised stream channels. a cone or fan builds out as the material is dropped. dry or mudflow fans formed by ephemeral streamflow.3 Alluvial Fans Alluvial fans are very dynamic landforms that can create significant hazards to highways as a result of floods. Good relationships exist between fan area and drainage basin area (Schumm 1977). Rejuvenated mature or old channels then exhibit some properties of youthful channels such as channel incision and erosion processes. Two different conditions of fan morphology are observed on modern dry fans. 5. these processes cause the total floodplain to raise in elevation. and avulsion (Schumm and Lagasse 1998).2 Floodplain and Delta Formations Over time. the stream channel is lengthened and the slope is further reduced. even old streams are far from static. As the bed material and water reaches the flatter section of the stream. Also. are affected by changes in sea level. is sometimes attributed to tectonic activity or climate change. There is considerable similarity between a delta and an alluvial fan. In some instances.2.2. Hence.presents several examples of landform evolution. and drainage basin characteristics are not surprising. the coarser bed materials can no longer be transported because of the sudden reduction in both slope and velocity. Fans can be of two types. the highlands of an area are worn down. The streams erode their banks. The presence of fan-head trenches. Low flat valley land and floodplains are formed. Both result from reductions in slope and velocity and both tend to reduce upstream slopes. Alluvial fans. As they work across the river valley. 5. The first situation occurs when deposition is near the mountain front and the fan surface is undissected. are changed by delta formations or glaciation. In the context of a rivers age. These relationships among fan slope. like deltas. tectonic activities or other causes. The material that is eroded is utilized downstream to build banks and bars to further enhance the meandering process. deposition. As deltas build outward the up-river portion of the channel is elevated through deposition and becomes part of the floodplain. channel incision. An action very similar to the delta develops where a steep tributary enters a main channel. Old rivers meander.

Above and below the profile break the river may be stable. which contrasts with dry fans. stored sediment has been flushed from the drainage basin. Normally the coarsest material is found at the fan apex. although fan-head trenching might result in a slight increase in sediment size with the fan radius. The first is due to intermittent uplift of the mountains which gradually steepens the fan head. and the second. a reconnaissance of the fan and its drainage should be undertaken so that potential changes can be identified and countermeasures taken. That is. where drainage basins that produced debris flows in 1983 do not contain sufficient stored sediment to produce debris flows at present. has a gradual change in elevation over a greater length of channel. The almost random distribution of erosion and deposition patterns on the arid fan is often replaced by a progressively shifting channel. 5. a dramatic change in channel morphology and stability occurs.2. which suggests very high sediment delivery from the drainage basin may. This condition can be identified in the field by observation or by the surveying of cross-fan profiles (Schumm and Lagasse 1998). Geomorphic thresholds controlling fan growth are sketched in Figure 5. Threshold concepts must be considered when evaluating fan-related hazards to highways. especially during floods. in fact.5). Therefore.2. Fluvial (wet) fans can become very large.3 . These perturbations are of two types: the first is a sharp break in profile which forms an in-channel scarp called a headcut (Figure 5. 5.4 Nickpoint Migration and Headcutting Abrupt changes in the longitudinal profile of the stream are shown in Figure 5.e.. being interrupted by periods of incision.2b). and it may be a very long time before sufficient sediment accumulates again to produce debris flows even under extreme rainfall. be an indication of future stability.1 (see also Section 5. and channel blockage and channel incision are important for highway design. Experimental studies show that growth at the fan-head is intermittent. This break in the profile induces a perturbation moving upstream. As the perturbation migrates past a point. Lowe 1993). topographic lows or swales). For example. deposition. called a nickpoint. Lateral migration of streams on fluvial fans can be anticipated by the concavity of the contours (i.2a). Two types of concavity are recognized. but still represents an oversteepened reach with respect to the overall channel slope (Figure 5. The other case is due to trenching and the building out of a low flatter reach of recent alluvium at the toe of the fan. This situation has been documented along the Wasatch Mountain front north of Salt Lake City (Keaton 1995. identification of relatively recent debris flow deposits. For example.2. The ideal result of any study of alluvial fans is a geomorphic map delineating active and inactive portions of the fan and the identification of problem sites within the active portions of the fan. To minimize these impacts on highways. not only the fan itself. The greatest variation in sediment yield is related to fan-head trenching and aggradation. but its drainage basin requires investigation.The longitudinal profile of fans may be concave. New orientation of a river channel is also an equally possible shifting process. The potential for avulsion. sediment reworking and downfan distribution of sediment. local aggradation in a channel can lead to avulsion because avulsion is likely to occur in places where deposition has raised the floor of the channel to a level that is nearly as high as the surrounding fan surface.

Changing slope at fan-head leading to fan-head trenching (Schumm 1977). California. scour continues until the gradient has been reduced and bank erosion has widened the channel to the point that deposition can begin. The result of headcut or nickpoint formation and migration is. Erosion of the bed material will be dramatic as a headcut or nickpoint migrates under a bridge. During a major flood in Tujunga Wash. lowering of the stream bed.Figure 5. As the nickpoint or headcut migrates farther upstream. Headcuts and nickpoints.4 . Figure 5. 5. erosion above the headwall of a gravel pit led to the failure of three highway bridges. a period of degradation may be followed at a site by a period of aggradation. In alluvial material.1. however a nickpoint produces degradation that persists for some distance (Figure 5. the quantity of sediment delivered to the reach at which a stream crossing is located increases greatly due to the erosion of the bed upstream and subsequent erosion of the banks of the stream. In both cases. headcut migration normally lowers the channel abruptly to its new position. Therefore. In cohesive material or erodible rock.2b). of course.2.

whereas the channel below shows significant erosion. engineers face the same problems and they have been successful in developing flood control. it is possible to define a valley slope above which the valley floor is unstable. but often at great cost and with the need to continually maintain and repair structures and channels (Schumm and Winkley 1994). dimensions.5 Geomorphic Threshold Evolution of a drainage network and sediment production from drainage basins are very complex processes. For example. Today. On topographic maps of large scale the presence of a nickpoint or headcut is indicated by closely-spaced contours. in a given drainage area. navigation. Geomorphic history and climatic changes introduce a new degree of complexity into the response of watersheds and fluvial systems. This is due to channel lengthening and gradient reduction accompanying increases in sinuosity and delta size.1 illustrates the concept of geomorphic threshold). historic change and channel response on the Mississippi River and the Nile are surveyed to illustrate the range of variability and change to be expected on alluvial rivers. one event can trigger a complex reaction as the components of the system respond to change. and Euphrates.The most obvious way to identify nickpoints or headcuts is by the use of aerial photographs. headcuts are very easily recognizable because the upstream valley floor or channel is essentially undisturbed. Nevertheless. Large alluvial rivers have always played an important role in human affairs. events with high magnitude and low frequency may at times have only minor and local effect on the landscape. the impact of human activities can greatly alter the behavior. Experimental studies demonstrate that within a complex natural system. 5. All of the early great civilizations rose on the banks of large alluvial rivers such as the Nile. Stable valley floors above the threshold line are incipiently unstable and any major flood may eventually cause erosion and trenching of the alluvium stored in these valleys (Figure 5. A low width-depth ratio below the break in slope is an indication of scour and deepening of the channel. these areas produce major land management and conservation problems. Indus. 5. Bank erosion is also a possible consequence and a sharp change in the bankline characteristics representing a change from stability to instability may identify the presence of a nickpoint or headcut. In this section. The magnitude of this complex response is likely to appear during early stages of an erosion cycle or during rejuvenation of high sediment producing areas. A change in the dimension of the channel and a change in the character of the bank line may indicate the presence of nickpoint or headcut migration. Since permanent changes result only when a geomorphic threshold has been exceeded. Yellow. and channel stabilization programs.2.5 . but at other times may produce seemingly "instantaneous" change with potentially catastrophic consequences. Particularly in arid and semi-arid regions. hence their practical interest. River engineering began early in human history to minimize the effects of floods and channel changes. It is possible to explain variations in sediment yield and channel adjustment by using the concept of geomorphic thresholds described by Schumm (1977). and general morphology of a river.3 VARIABILITY AND CHANGE IN ALLUVIAL RIVERS Those who work with rivers are aware of the great variability that exists among rivers and between river reaches. In addition. This can be verified with field surveys which show the break in the longitudinal profile of the stream. Another very common example of a geomorphic threshold is the progressive increase in channel sinuosity and meander amplitude until a cutoff or channel avulsion results on alluvial plains and deltas. 5.

For example. some streams are sensitive whereas others are not. care must be taken before the behavior and response of one stream can be extrapolated to another. The difference between these two great alluvial rivers. individual meanders frequently develop progressively to an unstable form. therefore. This significant lowering of base level caused incision of streams draining into the Mediterranean. as flume slope (and thereby sediment load and stream power) increased (Figure 5. the valley slope was steeper than was required to transport the reduced sediment load. In contrast to this unique history. 2. tributary contributions and other factors that affect valley slope. Stevens (1994) found that the Citanduy River in Java has remained relatively stable after a total of 23 cutoffs because it has very resistant clayey bed and banks. as a result of steepening of the gradient. The relatively straight River Nile and the relatively meandering Mississippi River reflect geomorphic history as well as current geologic controls such as active tectonics.3). Alluvial rivers differ in three ways: 1. An important consideration in predicting future river behavior and response is the sensitivity of the channel. many did not. Therefore. 5.6 . For example. Those that did not were characterized by stable banks and gentle slopes. which leads to local and short-term channel adjustments. will aid in predicting future river behavior and their response to human activities. to sinuous and finally to braided. The cutting off of numerous meanders along the Mississippi River caused dramatic changes. has much to do with their geologic and geomorphic history. and marine sediments are found as far up-river as Aswan. The Nile then filled this trough with fluvial sediments at a relatively gentle slope that was needed to move water and sediment to the newly established higher base level. A blockage at the Straights of Gibraltar stopped the inflow of Atlantic Ocean water into the Mediterranean basin. That is. marine water entered this canyon. especially the last two. although some responded in a manner similar to the Mississippi. Along any one river there can be considerable variability of channel morphology as a result of geologic and geomorphic controls. When sea level rose again. and geologic history (in other words. Information on these differences. Brice (1980) studied the effect of cutoffs and channel alignment on numerous smaller streams. and he found that. Obviously. an understanding of their complexity in space and through time is necessary. The differences among rivers is often reflected in their channel patterns. Rivers change naturally through time and as a result of climate and hydrologic change. However. sediment loads. how readily will it respond to change or how close is it to undergoing a change without an external influence? For example. Said 1981). the difference between the Nile and other large rivers may be due to the fact that the Mediterranean Sea evaporated during early Tertiary time (Hsü 1983. which led to serious bank erosion and scour (Winkley 1977).In order to cope with alluvial rivers. After retreat of the ice. the Mississippi River developed a relatively steep valley slope as the result of the influx of outwash sediment from melting continental glaciers. Rivers differ among themselves depending on hydrology. and a chute or neck cutoff results. rivers differ among themselves). and the Mississippi developed a sinuous course in order to reduce its gradient. During an experimental study of a channel in a laboratory flume. and a deep canyon was cut which formed the Nile valley. 3. the channel changed from straight.

(B) transitional meandering-braided channel with well-defined thalweg. (A) braided channel. 5. Figure 5.3.3.Figure 5. (D) relatively narrow and deep moderately sinuous channel. Relationship between flume slope and sinuosity (ratio between channel length and valley length) during flume experiments at constant water discharge (from Schumm and Khan 1972). This metamorphosis reflected great changes of discharge and sediment load. In addition to the Mississippi River. (C) low-sinuosity channel.4). many rivers completely changed their morphology and behavior (Figure 5. 5. Sequence of channel changes as water discharge and sediment loads decrease.1 Differences Through Time During the climate changes of Quaternary time. rivers on the Polish Plain (Kozarski and Retnicki 1977) and the Riverine Plain of Australia (Schumm 1968) underwent the same type of changes.4. and (E) multiphase meandering channel (after Schumm and Brakenridge 1987).7 .

Floods were reduced. Most meandering rivers will show significant change of sinuosity through time. and braided rivers could become sinuous. However. In fact. However.6). and cutoff. 5. grow. the type of response will depend upon the nature of the river. or the changes could be very minor. The Mississippi River was straightened and shortened 229 kilometers (142 miles) between 1933 and 1942. as river bends form. depending upon the sensitivity of the river (Schumm and Beathard (1976).5 shows the range of sinuosity between 1765 and 1930 for 24 reaches of the lower Mississippi River.3. 1994). Illinois and Old River. These are the types of changes to be expected with time along any active river. flood peaks. This frequently occurs on deltas. and formation of another. Figure 5. If as a result of climatic fluctuations and human activities. The greatest modern river changes have been the result of human activity. and it is exemplified by great lateral shifts of the Indus River (Holmes 1968). Figure 5. For example. 1994). and alluvial plains. Variability of sinuosity between 1765 and 1915 for 24 Mississippi River reaches. and water discharge of a river are altered.Perhaps of more interest here are the changes during shorter periods of time.8 . a river response can be expected. For example. Numbers identify the reaches of Figures 5. the sediment load.5. The reaches were identified by changes of valley slope. sinuous rivers could become straight. This great river experiment yielded both beneficial and undesirable results.2 Differences Between Reaches Perhaps of greater interest to river engineers is the variability along a single river. avulsion of the Mississippi River down the Atachafalaya River channel is only prevented by major flood-control structures. The goal was to reduce flood peaks and to improve navigation. but channel stability decreased (Winkley 1977. other factors intervene to cause considerable variability. One could assume that large alluvial rivers should have a relatively uniform morphology because the controlling factors of water discharge and sediment load should not vary greatly. 24 distinct reaches were identified between Cairo. a glance at even a coarse-scale map of the pre-cutoff Mississippi River reveals great variability. but the changes also can be avulsive and of a catastrophic nature with abandonment of one channel. Obviously the Mississippi River is composed of very different reaches that behaved differently during the period of record. Indeed.6 and 5.7 (after Schumm et al. 5. a distance of 768 valley kilometers (477 miles) (Figure 5.

The Mississippi is characterized by a lack of uniformity in its morphology and dynamics. Mississippi River reaches between Cairo. which has made the job of the river engineer difficult. Figure 5. Figure 5. Mississippi River valley profile (after Schumm et al. 1994). 1994).6.channel variability through time and the pre-cutoff 1930 river pattern. The differences among the reaches are for the most part the result of tectonic deformation of the valley floor and the presence of more resistant materials in the bed and or banks.7) shows considerable variability. Louisiana (after Schumm et al. A plot of the valley (floodplain) slope. Pleistocene-age alluvium. Figure 5. Illinois and Old River. 5. as a result of these controls. based upon the 1880 bankfull elevations (Figure 5. such as clay plugs. and Tertiary-age bedrock. 1994). The great variability is largely the result of geomorphic and geologic controls (Schumm et al.7.9 .5 shows great differences in the variability of sinuosity among reaches and for each of the reaches through time.

Several factors account for the fact that degradation has been minimal after closure of the dam.9) shows differences similar to that of the Mississippi (Figure 5. but because of the great population density along its course even minor changes are of importance. when studied using the same geomorphic techniques (Schumm and Galay 1994) is found to be much less variable and dynamic (Figure 5. Based upon an analysis of borings in the Nile valley. The change of valley slope and sinuosity below Qena is undoubtedly the result of the large Wadi Quena. and the bed of the Nile is sand.10). For example. and valley slope actually increases downvalley (Figure 5. Because degradation has been controversial. There appears to have been little change of the pattern of the River Nile since th mapping in the 18 century. Perhaps of greatest concern was the potential for major degradation of the Nile following construction of the dam.7). Mostafa 1957. Shalash 1980.10 . and such sediments undoubtedly were moved into the Nile during wetter periods. but they are much less. For example. and after.3 ft).5 mi) downstream from Armant." It is well known that only a small percentage of coarse bed material can armor a bed and it appears that the minimal degradation by the River Nile in response to the High Aswam Dam is the result of coarse sediment beneath a veneer of sand. Attia (1954) concluded that within the valley "coarse deposits composed of coarse sand. Sinuosity does increase with increased valley slope.5 to 28 ft). The anticipated impact of the High Aswam Dam on hydrology and sediment loads was a matter of great concern for engineers concerned with bank stability and potential channel degradation.70 m (2. However. such as faulting at Assiut. it is important to recognize that during past humid periods in Egypt. the wadis probably contributed abundant coarse sediment to the river. construction of the dam (Fathy 1956. gravel. and boulder size.Another great alluvial river. The other changes of slope are probably related to tectonics. A plot of the Nile valley slope (Figure 5. cobble. and even today wadi flooding must introduce coarse sediments into the river. All of the wadis contain much stored sediment of gravel.0 to 8. Boulders and cobbles are abundant in the trench. and El-Moatassem and El-Mottaleb 1979). the Nile.8). At present. maximum degradation was 0. or gravel lie beneath the fine alluvial deposits. 5. Gasser et al. wadis delivered coarse sediments to the valley. it was studied by many researchers prior to. Pre-dam estimates of degradation ranged from 2. A deep trench was excavated in a small wadi that enters the Nile valley at Khuzam about 30 km (18. during wetter periods of the past. except in the reach between El Saff and Cairo. and cobbles in Wadi Qena. there is an abundant supply of sand. many wadis appear to contribute only relatively fine sediment that can be readily transported downstream. which in the past has introduced coarse sediments into the Nile Valley. which could act as a control of degradation depending upon their location in the valley and the depth at which they are encountered. sand and gravel. Shalash 1983. which drains a large area to the north of Qena. but 18 years after the dam was in operation. Richardson and Clyma 1980. The construction of the High Aswam Dam commenced in 1963 and proceeded to 1968.5 m (6. 1978.

8 (after Schumm and Galay 1994).11 . Figure 5.8.Figure 5.9. River Nile valley profile. River Nile between Qena and Cairo showing six reaches of steep valley slope. Numbers identify reaches of Figure 5. 5.

Plot of River Nile sinuosity and valley slope. both the past and the present must be combined in what could be termed a geomorphicengineering approach to river maintenance and/or restoration. Therefore. they differ greatly morphologically and dynamically as a result of quite different histories. factors that are normally beyond the engineer's range of expertise may dominate river morphology and behavior.3.10. one cannot limit a study to modern conditions only.9. Both the Mississippi and Nile Rivers have sand-beds. 5.4 STREAM FORM AND GEOMETRY OF ALLUVIAL CHANNELS A study of the plan and profile of a stream is very useful in understanding stream morphology. although both the Mississippi and the Nile are large and transport similar sediments. and the Nile has apparently developed an armor of gravel and cobbles beneath its sand bed.12 . In order to understand alluvial rivers. 5. past channel morphology and behavior can provide valuable information about modern river behavior. Planview appearances of streams are varied and result from many interacting variables. Anomalous river behavior can often be explained by factors that are related to river history.Figure 5. Numbers represent reaches of Figure 5. in order to anticipate and to predict river response. Hence. yet what lies beneath the sand may be of critical importance to the character and response of these great rivers. For example. but it too responds to geologic controls and tributary influences. The vertical line joining two points for Reach 12 shows the difference between 18th century and present low sinuosity (after Schumm and Galay 1994). It is probable that the Mississippi is controlled totally by Tertiary-age clay and Pleistocene-age gravels.3 Summary The contrast between the lower Mississippi River and the Nile is striking. Both the Mississippi and the Nile provide support of this fact. The Nile is much less dynamic. In fact. Small 5.

the river is usually mature. meandering.12a.12c. In general. Levees form during floods as the river stage exceeds bankfull conditions. On the lower Mississippi River. Bed material transport is directly related to stream power.4. This classification is also used in HEC-20 (Lagasse et al. when the river valley is narrow and confined by terraces or valley walls. The rate of growth of natural levees is slower after they reach a height equal to the average annual flood stage.12. adversely affecting the river environment. 2001) as a basis for identifying geomorphic factors important to stream stability analyses. the highway crossing or encroachment can inadvertently change the planview or profile. Natural levees are a characteristic of alluvial river systems. It shows that straight channels are relatively stable only where flow velocities and sediment load are low.. the channel becomes braided. flow meanders in the channel causing the formation of alternate bars and the initiation of a meandering channel pattern. At high values of these variables. An example is given in Section 5. Farther from the river the gradients are flatter and the finer materials drop out.12b.1 Classification of River Channels Brice and Blodgett (1978) developed a simple classification scheme oriented primarily toward lateral stability of rivers.9 (Problem 2) on how to use this classification. Classification based on natural levees is illustrated in Figure 5. adversely affecting a highway crossing or encroachment. braided) are shown in Figure 5. flood control works and other river structures.13 shows the relative stability and types of hazards encountered. A channel classification in Figure 5. In this section. The common geomorphic terms for the various types of streams (e. The floodplain that is broad in relation to the channel width is indicative of an older river. Figure 5. The natural levees near the river are rather steep because coarse material drops out quickly. the silt and clay are essential to the growth of vegetation because of their water holding capability. The sinuosity index is the ratio of the length of the watercourse over the valley length between the same points. Well developed levees are associated with older rivers. Sediment is then deposited on the floodplain due to the reduced velocity and transporting capacity. By studying scroll formations in terms of age of vegetation the rate and direction of channel migration can be quantified. meandering channels are progressively less stable with increasing velocity and bed load.13 is also useful in making a qualitative assessment of stream stability based on stream characteristics. types of meander scroll formations are shown. stream form is classified and channel processes are discussed. In Figure 5. Classification based on oxbow lakes is illustrated in Figure 5. A detailed knowledge of the hydraulic characteristics of different types of streams is of great value when dealing with the location of highway crossings and encroachment. Similarly. The presence and size of point bars and middle bars are indications of the relative lateral stability of a stream channel.changes in a variable can change the planview and profile of a river. With good drainage. Each term is defined on the small sketches.13 . Beyond the levees are the swamp areas. 5. the growth of vegetation (tree cover) is indicative of the presence of silts and clays in the river banks and the floodplain. This is particularly true if the floodplain is well drained. and relative stability decreases as stream power increases as shown by Figure 5. 5.g. Additional information on specific channel features is illustrated in Figure 5. training works. Conversely. Conversely. As these variables increase. natural levees on the order of 10 ft in height are common.13.11.

Figure 5. Stream properties for classification (after Brice and Blodgett 1978).11.14 . 5.

15 .12. 1967).Figure 5. 5. Classification of river channels (after Culbertson et al.

16 . A knowledge of this behavior is important in anticipating and understanding stability problems.13. The next sections provide a brief discussion concerning the nature and stability of straight. 5. Further discussion of the opportunities and limitations related to using stream classification for river analysis. and Rosgen (1994 and 1996) are introduced in HEC-20 (Lagasse et al. Additional approaches to stream channel classification including those by Brice (1975). Schumm (1977 and 1981). Channel classification showing relative stability and types of hazards encountered with each pattern (after Shen et al. 1981).Figure 5. and meandering channels. 2001). Each behaves in a slightly different way when subject to human-related or natural impacts. Montgomery and Buffington (1997). braided. engineering and management can be found in Thorne (1997).

The thalweg oscillates transversely and initiates the formation of bends. Sloughing of the banks. Figure 5. high width-depth ratio. It is relatively active. 5. high energy river that has many bars.3 Meandering River Channels Alluvial channels of all types deviate from a straight alignment.14. that result in a sinuous thalweg (flow path connecting deepest points in successive cross sections) within the straight channel. and is relatively stable. The angle of deflection of the thalweg is affected by the curvature formed in the eroding bank and the lateral depth of erosion. has numerous bars and multiple thalwegs. thus. When the current is directed toward a bank.17 .4. the river engineer concerned with channel stabilization should not attempt to develop straight channels fully protected with riprap.5. nonuniform deposition of bed load caused by debris such as trees. In general. The second type is a steep gradient. The braided channel. and the Coriolis force due to the earth's rotation have been cited as causes for meandering of streams. In a straight channel the alternate bars and the thalweg are continually changing. as discussed in detail later.4.2 Straight River Channels Straight river channels can be of two types. and at low flow is braided. has a low width-depth ratio channel. the bank is eroded in the area of impingement and the current is deflected and impinges upon the opposite bank further downstream. Planview and cross section of a meandering stream. 5. The first forms on a low-gradient valley slope. the current is not uniformly distributed through the cross-section but is deflected toward one bank and then the other. The first type of straight channel may contain alternate bars (Figure 5.14).

local slope is steeper and velocities are larger in the crossing than in the pool. As a meandering river system moves laterally and longitudinally. or main current of the channel. flows from the pool through the crossing to the next pool forming the typical S-curve. As the point bars build out from the downstream sides of the bars. the location of the point bars. Years may be required before a configuration characteristic of average conditions in the river is attained. the bends gradually migrate down the valley.14. river slope approaches the valley slope at flood stage. the shifting of the current causes chute channels to develop across the point bar at high stages. The variability of bank materials. thalweg tends to straighten. In the crossings. Over time the local steep slope caused by the cutoff is distributed both upstream and downstream. Fine material that ultimately fills the bendway is plastic and cohesive. bends are formed by the process of erosion and deposition. 5. Point bars form on the inside of the bends. the sediment load may be sufficiently large to cause middle bars to form in the crossing. and the geometry of the adjacent bends. cause a wide variety of river forms in a meandering river.18 . In Figure 5. generally skewed in a downvalley direction. These plugs are sufficiently resistant to erosion to serve as semipermanent geologic controls and can drastically affect river geometry.14). Erosion without deposition to assist in bend formation would result only in scalloped banks. When a cutoff occurs. This causes the channel to appear as a bulb form. Erosion is greatest across the channel from the point bar. At higher stages. The lower end of the oxbow remains open and the drainage and overland flow entering the system can flow out from the lower end. At low flow the water surface slope is steep in the crossing and flatter in the pool. Overflow during floods carries fine materials into the oxbow lake area. Under these conditions the channel would simply widen until it becomes so large that the erosion would terminate. Figure 5. the channel cross-section is somewhat triangular. Usually the upstream end of the lake fills quickly to bank height. Note that in the crossing the channel is shallow compared to pools and the banks may be more subject to erosion. The bar is generally streamlined and its largest portion is oriented downstream. If there is very rapid caving in the bendways upstream.14 illustrates the change in water surface profile from low to high water discharge. shortening the path of travel and increasing the local friction slope. the channel cross-section is more rectangular and depths are smaller. The material eroded from the bank is normally deposited over a period of time on the point bars that are formed downstream. the meander limbs move at an unequal rate because of the unequal erodibility of the banks. As the river channel meanders it encounters old bendways filled with cohesive materials (referred to as clay plugs). The reverse is true at higher discharges. The point bars formed in the bendways clearly define the direction of flow. one can observe the position of the thalweg. thalweg is located very close to the outside of the bend. developing chute channels and a steeper channel prevails under this condition. At low flows. In the extreme case. The thalweg. Oxbow lakes may persist for long periods of time before filling. thalweg moves away from the outside of the bend encroaching on the point bar to some degree. At higher discharges the thalweg straightens. that is. alternate bars and the location of the pools and crossings. It is during high floods that the flow often cuts across the point bars. The oxbow gradually fills with fine silts and clays. In the extreme case. the bank material.The meandering river consists of pools and crossings (Figure 5. accounting for the lateral and longitudinal migration of the meandering stream. In general. In the pools. The channel geometry depends upon the local slope.12a). and the fact that the river encounters such features as clay plugs. At low stages. an oxbow lake is formed (Figure 5. The point bars constrict the bend and enable erosion in the bend to continue.

The actual meanders in natural rivers are obviously not as regular as indicated in Figure 5.In summary. meandering rivers have relatively flat slopes. and in many cases. Much of the sediment eroded from the outside bank is deposited in the crossing and on the point bar in the next bend downstream. The scour in the bend causes the bend to migrate laterally and sometimes downstream. A. and (5) bend deflection angle φ. (2) meander width W m. Definition sketch for meanders. sediments are deposited in this region. Point bar building is enhanced when large transverse velocities occur. and the bank-full channel width are shown in Figure 5. Because velocities are lower on the inside of the bend. large sandbars form in the crossings if the channel is not well confined.9 (Problem 1).15.16 and Table 5. The precise measurement of meander dimensions is therefore difficult in natural channels and tends to be subjective. In so doing. (3) mean radius of curvature rc. In the bends. It consists of a series of bends connected by crossings. The meander belt formed by a meandering river is often fifteen to twenty times the channel width.19 . These variables are shown in Figure 5.500 ft) per year have been observed in alluvial rivers. The bends are connected by crossings (short straight reaches) which are quite shallow compared to the pools in the bendways. 5. The geometry of meandering rivers is measured quantitatively in terms of: (1) meander wavelength λ. The centrifugal force in the bend causes a transverse water surface slope. deep pools are carved adjacent to the concave bank by the relatively high velocities. An example on how to measure these characteristics is presented in Section 5. Figure 5. At low flow. (4) meander amplitude A. they sweep the heavier concentrations of bed load toward the convex bank where they are deposited to form the point bar. forming a point bar. a meandering river has regular inflections that are sinuous in plan.1.15. Lateral movements as large as 760 m (2. helicoidal flow occurs in the bend. Some transverse currents have a magnitude of about 15 percent of the average channel velocity. The analysis of the mean meander dimension in nature shows that the meander length and meander width are both related to the width of the channels.15. When compared to most braided rivers. The empirical relationships for the meander length λ and the bank-full channel width as well as the meander amplitude.

Table 5. Figure 5.01 1. Multiple channels are generally formed as bars of sediment are deposited within the main channel. deposition occurs.Figure 5. Another cause of braiding is easily eroded banks.11 and 5.9W Inglis (1949) 1. As the channel steepens.1.9W λ = 4. Generally.6W λ 1.99 A = 18. and multiple channels develop.16.4 Braided River Channels A braided stream is one that consists of multiple and interlacing channels (Figures 5. In general.20 . If the channel is overloaded with sediment. the stream widens at high flow and forms bars at low flow which become stabilized.7rc Wolman (1960) 5. Meander Length to Amplitude to Meander Length to Channel Width Channel Width Radius of Curvature Source 0. a braided channel has a relatively steep slope.99 0.4. the bed aggrades. These interlaced multiple channels cause the overall channel system to widen. the velocity increases. Empirical relations for meander characteristics (Leopold et al. and the slope of the channel increases in an effort to obtain a graded state. a large bed-material load in comparison with its suspended load.98 Leopold and A = 2. 5. 1964). If the banks are easily eroded. thus forming islands.04 A = 10.10 0. and relatively small amounts of silts and clays in the bed and banks.13).6W Inglis (1949) = 6. the magnitude of the bed load is more important than its size.7W λ = 10. One cause of braiding is the large quantity of bed load. Empirical Relations for Meanders in Alluvial Valleys.17 will assist in defining the various conditions for multiple channel streams.

sediment discharge or both can cause significant changes in channel slope (see Section 5.21 .25 ≥ k (5. when: S Q 0. Similarly.18a and b illustrate the dependence of sand bed river form on channel slope and discharge. The changes in sediment discharge can be in quantity. is very wide and shallow even at flood flow and is. D50.2) 5. or both. They show that when: S Q 0. Often. in general. carries large quantities of sediment. 5. According to Lane (1957). Figures 5. Types of multi-channel streams. The braided stream may present difficulties for highway construction because it is unstable.5).1) where: k k = = 0. Qs.0007 SI 0.17.5 River Conditions for Meandering and Braiding It can be shown that changes in water discharge.4.Figure 5. changes its alignment rapidly.25 ≤ k (5.0017 English a sandbed channel meanders. such changes can alter the planview in addition to the profile of a river. unpredictable. or sediment size.

01 (a) LEO PO LD & WO LM LANE AN BRAIDED SLOPE (m/m) 0.00001 1 10 100 1000 10000 100000 MEAN ANNUAL DISCHARGE (m3/s) (b) LEO LANE 0.22 .0001 MEANDERING .01 25 SQ 0.4 4 = 0. 060 0.b.0001 MEANDERING SQ 0.001 LANE TRANSITIONAL SQ 0.18a.0.25 = 0.25 SQ 0.001 LANE TRANSITIONAL SQ 0.01 POL D &W OLM A BRAIDED N SLOPE (ft/ft) 0. Slope-discharge relationship for braiding or meandering in sandbed streams (after Lane 1957) (a = English units b = SI units). 5.01 0 SQ 0.4 4 = 0.0 041 =0 0.0 0070 0.25 = 0.25 = 0.0017 0.00001 100 1000 10000 100000 (ft3/s) 1000000 MEAN ANNUAL DISCHARGE Figure 5.

Leopold and Wolman (1960) plotted slope and discharge for a variety of natural streams. rivers. Leopold and Wolman recognize that their analysis treats only two of the many variables affecting morphology. but regard the width as a variable dependent on mainly discharge. alluvial streams do exhibit some quantitative hydraulic geometry relations. 5. Between these values of S Q is the transitional range.where: k k = = 0.025 the mean discharge in m /s (ft /s).3 uses these concepts in an engineering geomorphic analysis.010 English the sandbed river is braided.S. They further note that braided and meandering streams can be differentiated based on combinations of slope. two or more channels. plot in this zone. discharge.5. they do not expect this method to apply in every condition. these relations apply to channels within a physiographic region and can be obtained from data available on gaged rivers. because the data were all taken from natural streams. S is the channel slope in m/m (ft/ft) and Q is 3 3 0. It is understood that hydraulic geometry relations express the integral effect of all the hydrologic.6 Hydraulic Geometry of Alluvial Channels Hydraulic geometry is a general term applied to alluvial channels to denote relationships between discharge Q and the channel morphology.3) where: S Q k k = = = = Slope in m/m (ft/ft) 3 3 Bank-full discharge is m /s (ft /s) 0.23 . therefore.44 = k (5. In these equations.0125 SI 0. However. Many of the U. therefore. In alluvial channels. hydraulic and sedimentation characteristics of the channel are determined by a large variety of factors. If a river is meandering but its discharge and slope border on the transitional zone a relatively small increase in channel slope may cause it to change to a transitional or braided river. They observed that a line could separate meandering from braided streams (Figures 18a and b). and because the analysis obviously does indicate a significant relation between slope and discharge. the morphologic. 5. Channels forming in their own sediments are called alluvial channels.0041 SI 0. The mechanics of such factors are not fully understood.5. meteorologic.9 gives an example of this type of prediction and Section 5. In general. and geologic variables in a drainage basin for in-bank flows. and width/depth ratio. the analysis should give a reasonably effective prediction of channel pattern if slope and discharge are known. classified as intermediate sandbed streams. Problem 1 in Section 5.4. The equation of this line is: SQ 0. Braided streams are those which have relatively stable alluvial islands and. Leopold and Wolman note that sediment size is related to slope and channel pattern. but they do not try to account for the effect of sediment size on the morphology of the streams. hydraulics and sediment transport.06 English Streams classified as meandering by Leopold and Wolman are those whose sinuosity is greater than 1. However.

that in both basins.20.4) (5. For example. width 5. k. yo. Kennedy 1895. the width and depth at-astation do not change very much in Basin A.The hydraulic geometry relations of alluvial streams are useful in river engineering. Note. Lacy 1930. r and exponents b. The distinction between at-a-station and downstream hydraulic geometry relations is illustrated in Figures 5. they used Qs. The former are called at-a-station relationships and the latter downstream relationships.10) and b+ f +m=1 (5.7) (5. From the continuity equation (Q = WyoV).24 . as the discharge increases at-a-station. Hydraulic geometry relations were developed by Leopold and Maddock (1953) for different regions in the United States and for different types of rivers. p. c.6) (5.19 and 5. V and Qs to the variation of discharge at-a-station. In general the hydraulic geometry relations are stated as power functions of the discharge: W = a Qb yo = c Q V=kQ f m j (5. j. That is. Because QT is not readily available. and (2) relating these variables to the discharges of a given frequency of occurrence at various stations in a drainage basin. The width to depth ratio is almost constant but the velocity increases.20 illustrate how the hydraulic relations at-a-station and in the downstream direction may be different from one basin to another. Figures 5.19 and 5. it is seen that a x c xk = 1 (5.5) (5. t.8) (5. as it must.11) Leopold and Maddock (1953) have shown that in a drainage basin. f. In Basin B the width to depth ratio decreases with an increase in discharge. The forerunner of these relations are the regime theory equations of stable alluvial canals (see for example. two types of hydraulic geometry relations can be defined: (1) relating W. m. the suspended sediment transport rate. the width changes very little but the depth increases significantly with discharge at-a-station. y in these equations are determined from analysis of available data on one or more streams. z.9) QT = p Q Sf = t Q n=rQ y z where: W yo V QT Sf n Q = = = = = = = Channel width Channel depth Average velocity of flow Total bed sediment load Friction slope Manning's roughness coefficient Discharge as defined in the following paragraphs The coefficients a. and Leliavsky 1955).

increases with discharge in the downstream direction. Leopold and Maddock (1953) show that the type of relations illustrated in Figure 5. but in Basin B. Variation of discharge at a given river cross section and at points downstream (after Leopold and Maddock 1953). depth changes very little in the downstream direction. Figure 5.19.25 .20 exist in natural rivers. At-a-station relations pertain to individual sites such as A or B. Downstream relations pertain to a channel (segment A-B) or drainage network for discharge of a given frequency of occurrence. Figure 5. depth. and velocity with at-a-station and downstream discharge variation (after Leopold and Maddock 1953). 5.20. Schematic variation of width.

46 0.46 S f ≈ Qb −0.3.3 z -. Table 5.00 S f ≈ Qb D 50 Sand bed Gravel bed 5. Their relations for non-cohesive gravel-bed rivers in the downstream direction at bankfull discharge (or constant frequency) are also given in Table 5.4 +1.13 .05 .S.3. Derived At-A-Station and Downstream Geometry Relationships. . m. PA . Average At-A-Station Relations b f Average values Midwestern United States .5 2. .40 y o ≈ Q 0.S.45 Ephemeral Streams in Semiarid U.26 .34 . Applications of these relationships are illustrated in Section 5.3. and y as reported by Leopold et al.The mean values of exponents b. either upstream or downstream from a selected station.12 .28 -.26 V ≈ Q 0.2 m .1 .26 .5 .36 Average of 158 Gaging Stations in U.2.41 Average Downstream Relations (bank-full or mean annual flow) b f Average values Midwestern United States .42 . At-A-Station and Downstream Hydraulic Geometry Relationships.40 y o ≈ Qb y o ≈ Qb W ≈ Q 0.2 . These values are based on an extensive analysis of stream data in the United States and are stream specific.09 j . . PA . derived hydraulic geometry relations for conditions at-a-station (variable discharge frequency) and in the downstream direction on the same stream at bankfull discharge (or constant frequency) are given.43 j 2.41 Ephemeral Streams in Semiarid U.95 y -.08 V ≈ Qb 0.29 .9 (Problems 5 and 6). f.55 . Bray (1982) and Julien and Simons (1984) determined that bed material size is an important variable in hydraulic geometry relations in alluvial gravel streams.3 Appalachian Streams .46 0.8 1.07 . Table 5.40 Brandywine Creek.33 V ≈ Qb D0 50 −0. j.2 z .05 y -.S. Note: the term "downstream" implies any other location along the channel.04 .4 Brandywine Creek.55 .34 S f ≈ Q 0.3 In Table 5.36 m .33 W ≈ Qb D 50 0.45 Ten Gaging Stations on Rhine River .49 -1.53 −0. (1) (1) (2) At-A-Station Downstream Downstream 0. (1964) are given in Table 5.00 (1) (2) W ≈ Q 0. z.34 .07 -.5 .43 .2.

Work by Julien and Wargadalam (1995) has updated the theory and applications of hydraulic geometry relationships for alluvial channels. is examined from a 3-dimensional stability analysis of noncohesive particles under 2dimensional flows. 5. The downstream hydraulic geometry relations expressed in Section 5. it is important to attempt to verify this channel-forming discharge with field indicators of bankfull discharge. Appendix A to the 1999 USACE reference cited above provides a practical guide to effective discharge calculations. up to 90 percent of the total transport is caused by flows that are equaled or exceeded about ten percent of the time only. the average bed sediment load in a river may be described in terms of a formative discharge much larger than the mean annual flow. Thus.4.S. Research by the U. in terms of bankfull width.27 . Army Corps of Engineers (Watson et al. The data set covers a wide range of flow conditions from meandering to braided sand-bed. 1999. and gravelbed rivers.S. This discharge may be determined by integrating a sediment transport rating curve with the annual flow-duration curve. downstream hydraulic geometry of alluvial channels. Where possible. Copeland and Hall 1998) suggests that the effective discharge is the best representation of the channel forming discharge. depth and meander geometry may be defined in terms of different formative discharges rather than an arbitrarily chosen dominant discharge. Both the at-a-station and downstream hydraulic geometry relations are especially useful when the hydraulic design is based on the frequency of occurrence of flows. For arid region streams.6 relate to the bank-full stage. 5. rivers has a frequency of occurrence of one in 1.4. which for many humid region U. The analytical formulations were tested with a comprehensive data set consisting of 835 field channels and 45 laboratory channels. mean flow velocity. Four exponent diagrams illustrate good agreement with several empirical regime equations found in the literature. The concept of bank-full condition corresponding to a discharge with a period of return of 1.6 indicate how the channel morphology and other characteristics vary with discharge at-a-station or in the downstream direction in a drainage network.7 Dominant Discharge in Alluvial Rivers The hydraulic geometry relations discussed in Section 5. The concept of the frequency of occurrence of flows is important in the hydraulic design of highway crossings and encroachments. and friction slope. Analysis of bed sediment load estimations indicates that on most rivers.5 years for perennial streams and five to ten years for ephemeral streams is recommended for practical use when detailed analysis of formative discharge is not possible. average flow depth.5 years. the bankfull return period may be on the order of five to ten years. In their study. or feasible.4. The effective discharge is the increment of discharge that transports the most sediment on an annual basis. the relations need to be defined to determine the downstream hydraulic geometry of the channel at a site between two gaged sites. Also. the average channel width. The question then arises about the frequency of discharge to be used in the hydraulic geometry relations. In the hydraulic design of river crossings and encroachments.

5 QUALITATIVE RESPONSE OF RIVER SYSTEMS Many rivers have achieved a state of practical equilibrium throughout long reaches. This understanding can be obtained by: (1) studying the river in a natural condition. To do so. these stable reaches can be also called "graded" streams by geologists and "poised" streams by engineers. Application of Equation 5.5. this does not preclude significant changes over a short period of time or over a period of years. many streams contain long reaches that are actively aggrading or degrading.9 (Problem 7).4.8 River Profiles and Bed Material The slope of a river channel or a river system is usually steepest in the headwater regions. All too frequently the net result of a river improvement is a greater departure from equilibrium than that which originally prevailed. (2) having knowledge of the sediment and water discharge. However. Regardless of the degree of channel stability.12) where: Sx So α = = = Slope at any station a distance x downstream of the reference station Slope at the reference station Coefficient Similarly. 5. an understanding of the direction and magnitude of change in channel characteristics caused by human activity and natural processes is required. These aggrading and degrading channels pose a definite hazard to any highway crossing or encroachment. The river profile is concave upward and the slope of the river profile can be represented by the equation: S x = So e − α x (5. Generally. the bed sediment size is coarser in the upper reaches where the channel slopes are steep and the bed sediment size becomes finer with distance downstream.28 . Conversely. local human activities may produce major changes in river characteristics locally and throughout the entire reach. the size of the bed material reduces with distance according to the relationship D 50 x = D 50 o e − β x (5. (3) being able to 5. For practical engineering purposes. as compared to poised streams.13) where: D 50 x D 50o = = = Median size of bed material at distance x downstream of reference station Median size of bed material at the reference station Coefficient β The hydraulic geometry relations are applicable to continuous channel behavior. Good engineering design must invariably seek to enhance the natural tendency of the stream toward stable conditions.13 is illustrated with field data in Section 5. In some cases. this behavior (in this case the slope) may become discontinuous as the channel pattern changes from meandering to braided by the formation of cutoffs.

and 6. 3. methods have been developed to predict the response of channel systems to changes both qualitatively and quantitatively. Qs ~ τ o V W Cf D 50 (5. W is the width of the stream and Cf is the volumetric concentration of fine sediments. Shape of channel expressed as width-depth ratio is directly related to bed material discharge. bars. It is important to remember that these statements pertain to natural rivers and not necessarily to artificial channels with bank materials that are not representative of sediment load. and (4) applying a knowledge of geology. the data are not sufficient for quantitative estimates. soils. Examples of studies that have been undertaken by various investigators for qualitative estimates follow. Similar but more comprehensive treatments of channel response to changing conditions in rivers have been presented by Leopold and Maddock (1953). Because such a prediction is necessary.1 General River Response to Change Quantitative prediction of response can be made if all of the required data are known with sufficient accuracy. Lane (1955) studied the changes in river morphology caused by modifications of water and sediment discharges. These variables are interrelated and can respond to changes in a river system in the continual evolution of river form. Richards (1982). 1977). (1997). Meander wavelength is directly proportional to water discharge and to bed material discharge.predict the effects and magnitude of future human activities. A large number of variables are involved in the analysis. ASCE (1983). the relations will help to determine the response of water conveying channels to change. Equation 5. hydrology. Thorne et al. To predict river response to channel modifications is a very complex task. Thorne (1998). and hydraulics of alluvial rivers. Depth of flow is directly proportional to water discharge and inversely proportional to bed material discharge.29 .14 can be written as: 5. Slope of stream channel is inversely proportional to water discharge and directly proportional to bed material discharge and grain size.5. Research results support the following general statements: 1. In any event. 5.14) Here τo is the bed shear stress. Sinuosity of stream channel is proportional to valley slope and inversely proportional to bed material discharge. Width of channel is directly proportional to water discharge and to bed material discharge. Schumm (1971. 4. and forms of bed roughness all change with changing water and sediment discharges. V is the cross-sectional average velocity. Bed material sediment transport (Qs) can be directly related to stream power (τoV) and inversely related to the fall diameter of bed material (D50). The channel geometry. and only qualitative estimates are possible. 2. and Knighton (1998). 5. Santos-Cayado (1972). however. Usually.

30 . This increases the sediment discharge in the + main stream from Qs to Q s . and using Equation 5. The river bed will aggrade from C to C'. or both. Figure 5. is considered constant and if the concentration of wash load Cf can be incorporated in the fall diameter. then the relationship can be expressed as: QS ~ Q s D 50 (5.22. This possible change in planform geometry must be considered if any structure is to be built between points A and C'. In a second example.16) which is the relationship originally proposed by Lane (1955). the construction of a dam on a river usually causes a decrease in sediment discharge downstream. It is seen from Equation 5.22. The line CA (indicating the original channel gradient) therefore changes with time to position C'A. it can be concluded that for a decrease in bed material discharge from Qs to − Qs .16 is very useful to qualitatively predict channel response to climatological changes. the line CA. The proportionality represented by Equation 5. D50. consider a tributary entering the main river at point C that is relatively small but carries a large sediment load (Figure 5. for a significant increase in the + sediment discharge ( Q s ) the channel gradient (S) below C must increase if Q and D50 remain constant. Changes in channel slope in response to an increase in sediment load at Point C. Upstream of the confluence the slope will adjust over a long period of time to the original channel slope. Referring to Figure 5. Two simple example problems are analyzed using Equation 5.21). γ.15) If the specific weight. but use of physical diameter is sufficiently precise for most qualitative analyses. except Lane used the median diameter of the bed material as defined by sieving instead of the fall diameter.16 and the earlier discussion. This change may induce a change in the channel morphology downstream of point C as the downstream reach may tend toward braiding. In Figure 5. 5. The fall diameter includes the effect of temperature on the transportability of the bed material.21.Qs ~ γ y o S W V C f γ QS C f = D 50 D 50 (5. the slope S decreases downstream of the dam (assuming discharge and sediment size remain unchanged). In a first example. river modifications.16 that.16. representing the original channel gradient.

For this type of analysis. Also upstream of the dam the grade would return to the original equilibrium grade but would be offset vertically by the height of the dam. The prediction of the magnitude of possible errors in flood protection design. The engineer is also interested in quantities in addition to qualitative trends. indicating a decrease in bed elevation and slope in the downstream channel with time. In many instances it is important to assess the effects of changes in water and sediment discharge on specific variables such as depth of flow.31 . we can use Equation 5. To quantify these changes. Thus small reservoirs (storage capacity small in relation to annual discharge) may cause degradation below the dam and then aggradation over a relatively short period of time. channel width. that. except for local scour at the dam.4. 5. A plus (+) sign signifies an increase in the value of the variable and a minus (-) sign signifies a decrease in the value of the variable. The letter B indicates an increase in the 1/4 product SQ and a shift toward a braided condition and the letter M indicates a reduction in 1/4 SQ and a shift toward the meandering condition (see Section 5. however. Note. No attempt is made here to determine whether or not the channel actually braids or meanders. characteristics of bed materials and velocity.4. However. this initial step is useful because it warns of possible future difficulties related to channel modifications and flood protection works. Changes in channel slope in response to a dam at point C. The geomorphic relation QS ~ QsD50 is only an initial step in analyzing long-term channel response problems. the response of a river system to changes in specific variables is given in Table 5. if the dam fills with sediment so that the incoming sediment discharge passes through. the grade line C'A would return to the line CA.5).15 as follows: Q S D 50 / C f ~ SV y o W (5. requires the quantification of changes in stage. knowledge of the future flow conditions is necessary. because of changes in stage with time.changes to C'A. it is necessary to be able to quantify future changes in the variables that affect the stage. Figure 5. In this respect.17) Using this form.22.

It is important to notice that Einstein. Referring to Table 5. Size of Bed Sediment and Wash Load. Colby.Table 5.5. With the increase in discharge the stability of the channel would be reduced. Whether or not the channel form changes would depend on the river form prior to the increase in discharge. changes in river form and changes in river cross section both at-a-station and along the river channel using the other relations presented above. if the value of Qs increases.5.1 and 5.4. The information presented in Chapter 4 can be used to determine the direction of change of hydraulic variables when sediment characteristics or discharges are varied. and width will increase. On the other hand. Change of Variables Induced by Changes in Sediment Discharge. energy slope.1.32 .16 could be used to redict changes in channel profiles caused by changes in water and sediment discharge. velocity.5. As an example. this prediction could be affected by changes in form of bed roughness that dictate resistance to flow. Equations 5.2 Prediction of Channel Response to Change In Section 5. and Manning's equations apply to a cross section or reach and differ from some of the available geomorphic equations that have been derived by considering 5. It is now possible to talk qualitatively about changes in channel profile. Relationship + Qs + + − yo + D 50 / C f ~ S V / Cf ~ S V + W W Tendency to Braid or Meander B M B M M B B M B M B − + Qs D 50 / C f ~ S− V − y o W− + Q s D 50 + − yo + − ± Q s D 50 / C f ~ S− V ± y o W± ± Q s D 50 / C f+ ~ S− V ± y o W± Q s D 50 / − Qs − D 50 C f− ~S V − + + − yo W W + + + − Qs D 50 / C f ~ S+ V + y o W+ / Cf ~ S V ± ± yo − + + − Qs D 50 / C f− ~ S+ V + y o W+ − − − Qs D 50 / C f+ ~ S− V ± y o W± + Qs + D 50 / C f+ ~S V + + − yo W + Note: An increase in the value of the variable is denoted by a +. which indicates an increase in velocity. 5.25 and river regime. the slope. This can be best illustrated by application. it was illustrated that the proportionality of Equation 5. in the first line. The increase in discharge may affect the river form. cross-sectional area 0. consider the effect of an increase in discharge indicated by a plus sign on line (a) opposite discharge. the depth of flow will decrease and the channel may tend toward a braided form.18 (which illustrates SQ ) show that an increase in discharge could change the channel form in the direction of a braided form. and a decrease is denoted by a -. stability of the channel.2 or Figure 5.

increasing roughness and depth. and Manning's equations deal with depth of flow. and kinematic viscosity on the water and sediment discharge allows the establishment of the relative influence of those variables on stage-discharge relationships. the reverse is true. Also. Table 5. With seepage inflow. when blowing upstream. reducing the velocity. With more wash load. the fall diameter of the bed material is made smaller by significant concentrations of wash load. energy slope. The reverse is true with the wind blowing downstream. Seepage forces resulting from seepage losses help stabilize the channel bed and banks.5. The interdependency of top width. depth of flow. Wind generated waves and their adverse influence on channel stability are the most significant effects of wind. width of flow and energy slope whereas most geomorphic equations deal with channel depth. Colby. Variable Discharge Bed Material Size Bed Material Load Washload Viscosity Seepage Force Vegetation Wind (a) (b) (a) (b) (a) (b) (a) (b) (a) (b) (a) (b) (a) (b) (a) (b) Change in Magnitude of Variable + + + + + Outflow Inflow + Downstream Upstream Effect On Regime of Flow + + + + + + + + B→M M→B B→M M→B M→B B→M River Form M→B B→M M→B B→M B→M M→B Resistance to Flow ± ± + + + + + + + Energy Slope + + + + + + + + Stability of Channel + ± ± + ± ± ± ± + + Area + + + + + + + + Stage + + + + + + + + Chapter 4 indicates that the wash load increases the apparent viscosity of the water and sediment mixture. Einstein. In fact. bed-material size.33 . Wind can retard flow. 5. This information can be used to establish the direction of variation of hydraulic variables as a consequence of changes imposed on the bed-material and sediment discharge as shown in Table 5.a reach or total length of river. Vegetation adds to bank stability and increases resistance to flow. This makes the bed material behave as if it were smaller. channel width and channel slope. Qualitative Response of Alluvial Channels. the viscosity is affected by changes in temperature.5. the bed material is more susceptible to transport and any river carrying significant wash load will change from lower to upper regime at a lower Froude number than otherwise.

with or without an increase in channel depth. a straight river becomes sinuous and then eventually braided at high values of stream power and sediment transport (Figure 5. leading to increased bank erosion.4.18). channel deepening. All of these responses will cause some level of streambank erosion. Also. When only sediment load increases. The experimental work reported by Schumm and Khan (1972) shows that for a given discharge. Also. Changes in planform are most often exhibited through the downstream migration of meandering bends and changes in the sinuosity of meander bends. again.In terms of channel stability. Increased flow rates can result from an overall increase in the volume of water moving through the channel or an increase in channel slope. It is also important to recognize that the three geomorphic processes just discussed (channel widening. Generally. it is important to recognize that three geomorphic responses or processes can result from changes in dominant channel flow and sediment conditions. When both sediment discharge and flow increase. For example. adjustments in channel slope through degradation often are accompanied by increases in channel sinuosity and bank caving or channel widening. Rivers that are situated close to the meandering-braided threshold should have a history which is characterized by transitions in morphology from braided to meandering and vice versa. 5. An increase in flow or sediment discharge results in a tendency toward channel widening. Channel deepening is a process of channel degradation that increases the depth of the channel. however. For example.3). the channel section can be expected to increase its depth as well as its width. these changes are manifested by an adjustment of channel slope to conform with changes in flow or sediment discharge. a reduction in sediment discharge will result in an increase in channel sinuosity. Channel degradation can cause bank instability by producing a steeper bank angle. an aggrading channel reach can cause an increase in sinuosity in a downstream reach. the initiation of a given process at a particular site may initiate another process either upstream or downstream.5. Channel widening is evidenced through an increase in channel width. A reduction in sediment discharge or an increase in water discharge will result in a reduction of the channel slope. Other examples include the shifting of channels and the cutting off of meander bends. Whether or not instability actually occurs is a function of the properties of the bank materials and the original bank geometry. implying that the channel has filled in because of an excess of sediments. as valley-floor slope is progressively increased.3 River Pattern Thresholds and Response The work of Lane (1957) and Leopold and Wolman (1960) as summarized in Section 5.5 indicates that there is a gradient or discharge threshold above which rivers tend to be braided (Figure 5. Changing channel planform includes changes in channel pattern and position as viewed from above. and changing planform (a change in sinuosity or meander pattern). 5. channel deepening.34 . both of which lead to a tendency toward increased bank erosion. In this case the channel is said to be aggrading. and changing planform) are often interrelated and can occur simultaneously or in sequence. width increases but the depth may decrease. Channel deepening results from increased flow without an appreciable increase in sediment discharge. These are channel widening. These slope reductions result from increased channel sinuosity and/or channel-bed degradation.

The braided reach has a channel gradient of 0. because a braided river near the threshold might be converted to a more stable. However. The main channel is characteristically broad and shallow. the Chippewa River has a meandering configuration with a bankfull width of 194 m (636 ft) and a sinuosity of 1.49 1. If so. On the other hand. but there is a limit to the changes that can be induced beyond which the channel cannot function without a radical morphologic adjustment as suggested by Figure 5. the engineer can work with the river to produce its most efficient or most stable channel. Another approach might be to determine the position of the river on the Leopold-Wolman (1957) or Lane (1957) gradientdischarge graphs (Figure 5. The braided section has a gentler valley slope than the meandering reach upstream.00028.00035 0. The bankfull width as measured from U.0004 0.00035 as opposed to 0. Perhaps the best qualitative guide to river stability is a comparison of the morphology of numerous reaches. entering it 120 km (75 mi) below St. perhaps.00033 0.00033.6.00028 0. The Chippewa River rises in northern Wisconsin and flows 320 km (200 mi) to the Mississippi River.6).092 ft). single-thalweg stream. The sinuosity of this reach is very low.06. a meandering stream near the threshold should be identified in order that steps could be taken to prevent braiding due.28 Valley Slope 0.500 sq mi). The valley slope and channel gradient are different for each reach of the river.If the natural range of patterns along a river can be identified.4.S.00035 Channel Slope 0. Obviously. whereas the meandering reach has a gradient of 0. meandering. but the situation is reversed for channel slope.18). and the determination of whether or not there has been a change in the position and morphology of the channel in the past.49.092) 194 (636) 212 (695) Sinuosity 1. being only 1. If a braided river plots among the meandering channels or vice versa. a major tributary to the Mississippi River (Schumm and Beathard 1976). to changes of land use.00027 5.18. and it contains shifting sand bars. it is a likely candidate for change because it is incipiently unstable. Paul (Figure 5. Chippewa River Morphology.35 . An example of the way that this could be done is provided by the Chippewa River of Wisconsin. It is the second largest river in 2 Wisconsin. with a drainage basin area of 24 600 km (9.00040.5 mi) up the valley. then within that range it should be possible to identify the most appropriate channel pattern and sinuosity. Identification of rivers that are near the pattern threshold would be useful.06 1. 0. the Chippewa is braided (Table 5. Location Below Durand Above Durand Buffalo Slough Channel Pattern Braided Meandering Meandering Channel Width m (ft) 333 (1. a river can be forced into a straight configuration or it can be made more sinuous. Geological Survey topographic maps is 333 m (1. contrary to what is expected from Figure 5.5 km (16. From its confluence with the Mississippi to the town of Durand 26. in the 68 km (42 mi) reach from Durand to Eau Claire.23). Table 5.

provide a means of evaluating the relative stability of the modern channel patterns of the Chippewa River. The relations described by Leopold and Wolman (1957) and Lane (1957).36 . the river below Durand has remained braided during historic time. It has maintained its channel position and its pattern. The position of the braided reach as plotted on both figures indicates that this reach should be meandering. which resulted in a recent decrease in channel width of over 40 percent. This suggests that the braided reach is anomalous.18.23.Figure 5. but both are well within the meandering zone. When the Chippewa data are plotted on Lane's graph (Figure 5.082 cfs). The braided reach plots higher than the meandering reach. 5. but within the range of scatter about the regression line for meandering streams.416 cfs) is used. which is the flood discharge having a return period of 2. The only other change in the lower river is a noticeable growth of the Chippewa delta into the Mississippi valley. the braided reach still plots within the meandering region of Figure 5.503 3 m /s (53. but a significant narrowing as the result of the attachment of islands and the filling of chute channels has occurred downstream of Durand. that is. There is no evidence to suggest that the Chippewa River is either progressively eroding or aggrading its channel at present. Even when the 25-year flood of 3 2.787 m /s (98. according to this relation the lower Chippewa would be expected to display a meandering pattern rather than a braided one.24. Map of lower Chippewa River. The Chippewa River falls in the intermediate region.25) the same relation exists. Again the braided reach is seen to be anomalous because it should plot much closer to or above the braided stream regression line. which are combined in Figure 5.24 (the Leopold and Wolman relationship) for both the braided and the meandering reaches of the Chippewa. as defined by Leopold and Wolman.33 years. In fact. The value used for the bankfull discharge is 1. Wisconsin. The river is braided below Durand. The delta deposits show increased vegetational cover as well as extension into the Mississippi River valley. The bankfull discharge was plotted against channel slope on Figure 5. Buffalo Slough is an old course of Chippewa River (Schumm and Beathard 1976).

and mean discharge. The regression lines were fitted to data from streams that Lane classified as highly meandering to braided. Figure 5. channel gradient.25.37 . Letters B and M identify position of the braided and meandering reaches of the Chippewa River (Schumm and Beathard 1976). and bankfull discharge. 5.Figure 5. Lane's (1957) relation between channel patterns. Letters B and M identify braided and meandering reaches of Chippewa River. Leopold and Wolman's (1957) relation between channel patterns. Letters K and W refer to Kowhai and Wairau Rivers (Schumm and Beathard 1976). channel gradient.24.

28 is. The Rangitata River is the southern most of the major rivers which traverse the 5. The normal configuration of the lower Chippewa is sinuous. and indeed flow was completely eliminated in 1876. the present sediment load carried by the Chippewa River is greater than that conveyed by the Buffalo Slough channel. This sinuous channel could not have transported the large amounts of sediment that the present braided channel carries to the Mississippi River. Channel sinuosity decreased from 1. and the resulting bed and bank erosion has supplied large amounts of sediment to the Mississippi. the meandering pattern of the Buffalo Slough channel was appropriate.38 . there is a significant morphologic feature on the Chippewa River floodplain. The channel shifted from Buffalo Slough to a straighter. it is also the ratio of valley slope to channel slope. steeper course along the northwestern edge of the floodplain. A sinuosity of 1. more detailed studies of the Chippewa River basin indicate that upstream sediment production must be controlled. The sinuosity of Buffalo Slough is approximately 1. the ratio of the present valley slope (0. the lower Chippewa could resume its sinuous course. thereby maintaining a channel gradient of about 0. Flow through Buffalo Slough has decreased during historic time. but this is due almost entirely to the formation of the new straight. and it is evidence of a major channel change in the Chippewa River valley (Table 5. This more efficient route gradually captured more and more of the total discharge.49 to 1. the Buffalo Slough channel had a gradient that was not appreciably different from that in the upstream reach. when the upstream end of Buffalo Slough was permanently blocked.This conclusion requires an explanation that can be based on the geomorphic history of Chippewa River. The abandonment of the former Buffalo Slough channel by the Chippewa River is the result of an avulsion.0027.6). An appropriate means of channel stabilization and sediment load reduction in this case is the development of a sinuous channel. the high sediment delivery from the Chippewa might be controlled. The new Chippewa channel produced a braided configuration due to a higher flow velocity and the resulting bank erosion. This channel slope value is very similar to the channel slope of the meandering reach upstream of Durand (0.28 in a downstream direction. An indication that the pattern conversion of the Chippewa could be successful if the upstream sediment sources were controlled is provided by the Rangitata River of New Zealand. Since the above suggestions were made (Schumm and Beathard 1976). which occupies the southeastern edge of the floodplain (Figure 5.00028) therefore. It appears that the lower Chippewa has not been able to adjust as yet to its new position and steeper gradient. For example. but one that took many years to complete. If the contribution of sediment from these sources were reduced.00027. or it too would have followed a straight braided course.00035) to a channel slope of about 0. Although sinuosity is usually defined as the ratio of stream length to valley length. braided channel. and if it could be induced to assume such a pattern. As the new channel grew. the old course deteriorated and eventually its discharge was so reduced that the Mississippi was able to effectively dam the old channel mouth with natural levees and plug its outlet. therefore. Although delta construction at the Mississippi River confluence was responsible for the lower valley slope of the lower Chippewa River valley. especially where the upper Chippewa River is cutting into the Pleistocene outwash terraces.28. Buffalo Slough. steep. Therefore. which is 121 m (397 ft) wider than the old sinuous Buffalo Slough channel.23). It is a sinuous remnant of the Chippewa River that was abandoned.

the valley of the Rangitata is braided. and this was accomplished by the construction of curved training banks.6 MODELING OF RIVER SYSTEMS The necessity for quantitative prediction of river channel response is increasing. it appears that the pattern threshold can be crossed successfully. the engineers have had success in converting the Wairau River. high sediment loads are delivered to the Kowhai River directly from steep and unstable mountain slopes. the Rangitata is meandering. it is very difficult to accurately formulate mathematically what happens in a river. One is the mathematical model and the other is the physical model. In fact. the river cuts into high Pleistocene outwash terraces. because the Rangitata is obviously a river near the pattern threshold. Perhaps the simplest way to determine the most likely pattern for a river is to determine its pattern from the earliest maps and aerial photographs. to date they have been best used to study channel response using 1-dimensional. If the Rangitata could be isolated from the gravel terraces. Studies of channel 5. it probably could be converted to a single-thalweg sinuous channel.05. it seems unlikely that an attempt to convert to a sinuous channel will be successful. or at most 2dimensional. approximations. A few miles farther downstream. On Figure 5.24 and 5. it may be difficult to maintain a single-thalweg channel at this location. below the gorge. as are all the other rivers which cross the Canterbury Plain. However. The variability of the Rangitata River pattern indicates that braided to single-thalweg conversions should be possible for the Chippewa and Wairau Rivers. New Zealand engineers are attempting to accomplish this pattern change in order to produce 'single-thread' channels which will cause less flood damage. Mathematical models utilize a number of mathematical equations governing the motion of water and sediments in a channel.24 the Kowhai River plots well above the threshold line. although there have been no major erosional or hydrologic changes in the drainage basin.0 to 1. For complex three-dimensional channel processes.25.Canterbury Plains of South Island. and it appears that the Rangitata should be a braided stream below the gorge. have greater stability at bridge crossings. There are other New Zealand rivers that are near the pattern threshold. single-thalweg. The accuracy of such a prediction depends on the quality of the data. The increase in sinuosity is only from 1. relatively more stable channel. Regardless of the potential of mathematical models. However. and without a major reduction in upstream sediment. a major braided stream.24 the Wairau River plots close to the threshold line. Whereas much of the sediment load in the Wairau River is derived from bank and terrace erosion which can be controlled. and with the reduction of sediment load produced by bank stabilization. and be less likely to acquire large sediment loads from their banks and terraces. However. It leaves the mountains through a bedrock gorge. the Kowhai River is being modified in the same manner as was the Wairau. For example. from its uncontrolled braided mode to that of a slightly sinuous. The braided pattern persists to the sea. and it abruptly converts from a meandering to a braided stream. Above the gorge. If the river was braiding in the past. if the historic river was meandering and it is now braided. and therefore. 5. Farther to the south near Kaikoura. as this depends on their position with regard to the line defining the pattern thresholds on Figures 5. they are susceptible to pattern change.39 . There are generally two ways of predicting response. If the subject river is very different it may be converted to the local character if neither the sediment load or hydrologic character are significantly different. then a conversion back to a meandering pattern is appropriate. On Figure 5. not all braided rivers can be so readily modified. Another approach is to determine the characteristics of other nearby rivers.

and 2. Therefore. For kinematic similarity. time. To each point. The performance tests of proposed structures can be made at moderate costs and small risks on small-scale (physical) models. patterns or paths of motion between the model and the prototype should be geometrically similar. while the Reynolds number represents the ratio of inertial to friction (viscosity) forces. 2001) discusses the use of physical modeling in bridge scour and stream instability countermeasure design. the Froude criterion is used to determine the geometric scales. geometric. 5. If the two systems are denoted by the subscript m for model and p for prototype. If similarity of flow is maintained between the model and prototype. the prediction of a given phenomena in time and space domains can be made mathematically. departure from strict similarity. To satisfy the preceding conditions in clear water. gravitational forces predominate. The Froude number represents the ratio of inertial to gravity forces in the system being modeled. The relationship of the governing physical processes and parameters must then be the same for the model as for the prototype. while dynamic similarity is a scaling of masses and forces. The physical model is designed to achieve similar behavior as that of the prototype.6. mathematical equations of motion will be identical for the two. but only with the knowledge that some scale effects. the interaction of a structure and the river environment can be studied in detail. In such cases.response to development for complex situations are usually made using a physical model. dynamic and geometric similarity all at the same time in a model. It is seldom possible to achieve kinematic. kinematic. a uniquely coordinated point. time and process in the prototype. The ratios of corresponding physical magnitudes between prototype and model are constant for each type of physical quantity. Also. physical models are used to physically represent solutions to the governing equations.1 Physical Modeling Physical models are used to test the performance of a design or to study the details of a phenomenon. all the governing equations are not known. In this regard. in open channel flow. the effects of the Froude number are more important than those of the Reynolds number. Kinematic similarity refers to similarity of motion. Similarly. Considering the equations of motion. and hence. then 5. the known equations cannot be directly treated mathematically for the geometries involved. and dynamic similarities must exist between the prototype and the model. exists in the model. the dimensionless ratios of V / gy (Froude number) and Vy/ν (Reynolds number) are both significant parameters in models of rigid boundary clear water open channel flow. The natural phenomena are governed by appropriate sets of governing equations. that is. Ratios (or scales) of velocity. If these equations can be integrated. In many cases related to river engineering. Rigid Boundary Models.40 . force and other characteristics of flow for two systems are determined by equating the appropriate dimensionless number which applies to a dominant force. HEC-23 (Lagasse et al. Geometric similarity refers to the similarity of form between the prototype and its model. For instance. time and process exists in the model. The similitude required between a prototype and a model implies two conditions: 1.

7 for the Froude (gravity). Equation 5. may become important and complicate the interpretations of results of the model. The use of this table is presented in Section 5. A large length scale also ensures that the flow is as turbulent in the model as it is in the prototype. and Weber (surface tension) criteria. in free surface flow. the length ratio is often selected arbitrarily. For example. The Froude number is used as a scaling criterion because gravity has a predominant effect.41 . Reynolds (viscosity).9.18) for the coordinate directions.21) It is not always possible to achieve boundary roughness in a model and prototype that corresponds to that required by Equation 5.18 assumes a condition of exact geometric similarity in all coordinate directions. y. Frequently. 5.19) and zr = zm / zp (5. L r = xm / xp = y m / y p (5. and z. Thus. Analysis of Manning's equation and substitution of the appropriate length ratios. The boundary roughness is characterized by Manning's roughness coefficient. if a small length ratio is used (very shallow water depths) then surface tension forces. The subscript r is used to designate the ratio of the model quantity to the prototype quantity. may be necessary to offset disproportionately high resistance in the model. which are included in the Weber number ( V / σ / ρL ) . However. It is desirable that the length scale be made as large as possible so that the Reynolds number is sufficiently large that friction becomes a function of the boundary roughness (and essentially independent of the Reynolds number). results in an expression for the ratio of the roughness which is given by nr = L r 1/ 6 (5.21 and additional measures such as adjustment of the slope. geometrically distorted models can have different scales in horizontal (x.20) If perfect similitude is to be obtained the relationships that must exist between the properties of the fluids used in the model and in the prototype are given in Table 5.the ratio of corresponding quantities in the two systems can be defined. and two equations are necessary to define the length ratios in this case. A model is said to be distorted if there are variables that have the same dimension but are modeled by different scale ratios. Problems 5 and 6. x. based upon the Froude criterion. y) and vertical (z) directions. open channel models are distorted. the length ratio is given by: L r = xm / xp = y m / y p = zm / zp (5. n. In free surface flow. but with certain limitations kept in mind.

Distortions of various parameters are often made in such models. Similitude in mobile bed models implies that the model reproduces the fluvial processes such as bed scour. These models have the bed and sides molded of materials that can be moved by the model flow. The predictive use of the model should be restricted to the aspects for which the model has been verified.42 .and three-dimensional mobile bed models are often used. In general. Blench. and others (Mahmood and Shen 1971). the model scales need adjusting during the verification stage. One approach is the analytical derivation of distortions explained by Einstein and Chien (1956) and the other is based on hydraulic geometry relationships given by Lacey. and varying boundary roughness. bed deposition. Model verification consists of the reproduction of observed prototype behavior under given conditions on the model. a model may be verified for bed-level changes over a certain reach of the river. The use is based on the premise that if the model has successfully 5. It is generally not possible to faithfully simulate all of these processes simultaneously on scale models. a first approximation of the model scales and distortions can be obtained by numerical computations. In both of these approaches. The model is built to these scales and then verified for past information obtained from the prototype. This is specifically directed to one or more alluvial processes of interest.7. Scale Ratios for Similitude.Table 5. Two approaches are available to design mobile bed models. In modeling highway crossings and encroachments in the river environment. For example. two. lateral channel migration. Characteristic Length Area Volume Time Velocity Acceleration Discharge Mass Force Density Specific Weight Pressure Impulse & Momentum Energy and Work Power Dimension L 2 L 3 L T L/T 2 L/T L /T M 2 ML/T 3 M/L 2 2 M/L T ML/T 2 2 ML /T 2 2 ML /T ML /T 2 3 3 Re L 2 L 3 L ρL /µ µ/Lρ 2 2 3 µ /ρ L Lµ/ρ L ρ 2 µ /ρ ρ 2 2 µ /L ρ 2 2 µ /L ρ 2 L µ 2 Lµ /ρ µ /Lρ 3 2 3 2 Fr L 2 L 3 L (Lρ/γ) 1/2 (Lγ/ρ) γ/ρ L 5/2 1/2 We L 2 L 3 L (L ρ/σ) 1/2 (σ/Lρ) 2 σ/L ρ L 3/2 3 1/2 (γ/ρ) 3 1/2 (σ/ρ) L ρ Lσ 3 1/2 L ρ 3 L γ ρ γ Lγ 7/2 1/2 L (ργ) 4 L γ L 7/2 ρ 2 σ/L σ/L 5/2 1/2 L (ρσ) 2 Lσ σ 3/2 γ 3/2 ρ -1/2 (Lρ) 1/2 Mobile Bed Models.

S. and Channel Stability Analysis of the Schoharie Creek Bridge Failure. 5. many successful examples of their use are available. More information on physical modeling can be found in HEC-23 and reports by Gessler (1971). The mobile bed models are more difficult to design and their theory is significantly more complicated as compared to clear water rigid bed models. This is where numerical. J. which can simulate both short. important river training and control works are invariably studied on physical models. Bonner Bridge. (1987. carefully implemented numerical methods to obtain approximate solutions to the appropriate partial-differential equations. et al. These unsteady flow models have succeeded because they are based on mathematical descriptions that incorporate all the important physical processes involved and use reliable. D.reproduced the phenomenon of interest over a given hydrograph as observed on the prototype. 1999).M.43 . In general.M. Yalin (1971).and long-term response. the sheer expense and space requirements of physical scale models generally disqualify them for simulation of long-term. 1987)." (Richardson et al.2 Computer Modeling The design engineer's interest in alluvial river response is generally focused on anticipating how the river bed and water-surface elevations will change if an existing stable or equilibrium situation is perturbed. unsteady flood propagation models that have proven so useful in engineering design. Numerical models of alluvial river response are the natural outgrowth of rigid-boundary. "FHWA Hydraulics Laboratory and Partners Perform Scour Evaluation for Woodrow Wilson Bridge" (Jones. and "Local Pier Scour Model Tests for Jensen Beach Bridge. This perturbation may be the occurrence of an unusually large annual flood that temporarily scours the bed and banks to accommodate the higher flow before returning to normal conditions. this information is of great help in comparing the performance of different designs. The perturbation may also be a permanent change in river discharge and sediment supply caused by upstream regulation of flows. Shen (1979). 1999). or a change in channel geometry resulting from bank stabilization or channelization. 1996).6. Even in the many cases where it is only possible to obtain qualitative information from mobile bed models. However. "Flume Modeling Experimental Plan for the Replacement of the Herbert C." (Parsons Brinkerhoff Quade and Douglas. 1989). computer-based models. find their natural area of application. New York. However. "Laboratory Report of the Acosta Bridge Scour Study. alluvial river-response models have not enjoyed the success of their rigid-boundary cousins. Adoption of a particular method for estimating river response depends on quality and availability of data as well as the engineer's experience. 1990). The interpretation of results from a mobile bed model requires a basic understanding of the fluvial processes and some experience with such models." (Stein. it will also reproduce the future response of the river over a similar range of conditions. Although problems arise with interpretation of the results. However. precisely because of the weaknesses in our 5. large-scale river bed response to the second type of perturbation. Erosion. and Richardson et al." (Sheppard. Inc. physical models in the hands of experience modelers can yield valuable information on local scour and deposition around structures. Additional applications of physical model studies of alluvial channel flow at highway crossings can be found in: "Hydraulic. Another area where prediction of long-term response is desired is river stabilization and river restoration design. The first type of perturbation can often be simulated using a physical scale model. S.

two basic approaches are possible: coupled or uncoupled. Most published 1-dimensional software systems for the solution of the water. known inputs to the sediment continuity and transport equations. sorting. as long as it is consistent with the partial-differential equations and is stable. especially when the sediment-transport flow resistance equation involves not just an analytic mathematical expression. When the water wave propagation effects are of secondary importance for sediment-transport phenomena. Notwithstanding this fundamental difficulty. these equations then become relatively easy to solve numerically. velocities. because the water-flow and sediment-transport processes occur simultaneously. has only a secondary effect on simulation quality. mutually dependent processes. Whether the full unsteady set of equations or the quasi-steady set of equations is solved numerically. but in one dimension there does not appear to be any strong reason for doing so. and sediment-transport relationships. The uncoupled procedure has arisen essentially to circumvent the computational difficulties of the coupled approach. armoring. The uncoupling of the liquid and solid transport occurs during a short computational time step. the system of equations can be simplified by assuming that the water flow remains quasi-steady during a certain interval of time. and hydraulic research engineers have targeted alluvial river hydraulics as a prime area for continuing fundamental and applied research. These equations form a nonlinear partial-differential system that in general cannot be solved analytically. yielding the new bed elevations. these processes are simulated in a third uncoupled computational phase using new depths. and so forth. Although it is difficult to quantify the error associated with this artificial uncoupling of simultaneous. The particular numerical method used. and bed elevations as known inputs. but a whole series of procedures and computations to simulate alluvial channel processes such as armoring. conservation of sediment. design engineers have an immediate need for reliable numerical simulations. Out of this fortunate confluence of interest have arisen a variety of simulation techniques and software systems.and sediment-flow equations use one form or another of the finite-difference method. conservation of water momentum. However. Then the depths and velocities are taken as constant. When the overall model includes bed-sediment sorting or armoring. In any case. as well as many apparently successful simulations of prototype situations. Some authors have used the finite-element method. Experience in the use of uncoupled models. and bed forms. This is evidently the physically proper way to proceed. the quality and reliability of numerical models for bed evolution are determined primarily by the sediment-transport formulation and mechanisms adopted for sorting. assuming that neither the bed elevation nor the bed-sediment characteristics change during the time step. The most basic one-dimensional description of water and sediment flow in an alluvial river consists of four relations: conservation of water. a simultaneous solution of both water and sediment equations is sought.understanding and mathematical formulation of the relevant physical processes. it is intuitively obvious that the uncoupling is justified only if bed elevations and bed-material characteristics change very little during one time step. In the coupled case.44 . leading to a system of algebraic equations. the simultaneous solution may involve certain computational complications. in which time and space derivatives are approximated by differences of nodal values of grid functions that replace the continuous functions. First the water-flow equations are solved to yield new values of depth and velocity throughout the reach of interest. with both the unsteady 5.

alluvial riverbed prediction models that are readily available for application to prototype cases seems to be quite small. The BRI-STAR (Molinas 2000) model developed for the Federal Highway Administration is a semi-two-dimensional model capable of computing alluvial scour/deposition through subcritical. Numerical modeling of alluvial river flows has become very popular in the 1990s because of the advancement of computer technology. others ignore them completely. it should be recognized that the short-term models can also be applied for long-term prediction if variable time steps are employed. and flow resistance through both particle roughness and bed-form effects. river regime (existence of ripples and dunes). Many of the available models have been developed for specific rivers under particular flow and alluvial river bed conditions. and small boulders). The minimum rate of energy dissipation theory is used for decisions as to whether channel adjustments taking place at a given cross section. so that the material remaining on the bed contains a progressively higher proportion of coarser material. but become even more difficult to quantify. The stream tube concept is used in a semi-two-dimensional way.45 . an interlocking armor layer may form on the surface. The short-term models are best suited to compute changes in alluvial river bed level during a relatively short time period. Some models attempt to simulate sorting effects on bed evolution. BRI-STARS MODEL (The Bridge Stream Tube Model for Alluvial River Simulation). and many of them are. However. and a combination of both flow conditions involving hydraulic jumps.and quasi-steady water-flow equations. Both energy and momentum equations are used so the water surface profile computation can be carried out through combinations of subcritical and supercritical flows without interruption. from gravels and coarse sands down to fine silt and clay in varying proportions. thus affecting the sediment-transport rate. which allows the lateral and longitudinal variation of hydraulic conditions as well as sediment activity at various cross sections along the study reach. The sediment continuity equation and sediment transport capacity equations are used for sediment routing computations to simulate the general scour in the river bed elevation. Another distinguishing feature of numerical bed-evolution models is the representation of sediment sorting and bed-surface armoring. These processes are qualitatively reversed during deposition. to some extent. arresting further degradation. Finer particles are preferentially entrained into the flow as erosion occurs. A broad range of sizes are represented. Thus another important distinguishing feature of computer-based models is the degree to which they incorporate sorting and armoring effects. cobbles. However. Alluvial sediments are rarely of uniform grain size. The sediment routing is performed for each size fraction to account for the bed composition changes and the bed armoring processes. and thus are suited for long-term prediction of river bed level for multiple-flood events over multiple years. If the original bed material contains a high enough proportion of large. 5. Long-term models employ simpler implementations of steady-state flow equations. has shown that the uncoupling is not a serious obstacle to successful simulation. This so-called sorting process tends to increase the mean bed-sediment size as degradation occurs. a shorter time step is used for highly unsteady flows and a longer time step is used otherwise. BRI-STARS is capable of simulating channel widening/narrowing phenomenon. the number of computer-based. non-moveable materials (coarse gravel. supercritical. The following assessment of selected models is made for two different groups: short-term models and long-term models. well tuned or calibrated only for those particular rivers. In that case.

which makes possible a detailed accounting of hydraulic sorting and development of an armored layer. central-difference.S. and long-term trends of scour or deposition in a stream channel. FLUVIAL-12 first solves the unsteady. Width adjustments are made in such a manner that the spatial variation in power expenditure per unit channel length is reduced along the reach by a trial and error technique. with sediment-transport capacities determined from water-flow conditions previously determined in the uncoupled backwater computation. The entire movable bed portion of the channel is assumed to aggrade or degrade uniformly. the net change in cross-sectional area is obtained by solving the sediment-continuity equation using a backward-difference scheme for space and a forward-difference scheme for time. Next. unsteady. 5. The sediment-continuity equation is solved using an explicit finite-difference scheme. The model also includes the capacity to simulate bank erosion and is described as an erodible-boundary model as opposed to an erodible bed model by Chang. degradation of the streambed downstream from a dam. The computed cross-sectional area change is then adjusted for the effects of channel migration. The user can select from thirteen sediment transport functions including one general function including depth and slope. which incorporates the effect of radius of curvature of the river bend into the transverse component of the sediment transport rate. simulates the scouring/deposition process taking place in the vertical direction across the channel. and is intended for long-term river response simulation. including the effects of dredging. FLUVIAL-12. gradually varied.46 . It is anticipated that future versions of HEC-RAS will incorporate sediment transport and replace HEC-6. The model uses an implicit. water and sediment flows for channels. the total stream power minimization algorithm. The first component. The user can select from one of seven sediment transport functions including one general function including velocity depth. the bridge component allows the computation of the hydraulic flow variables and the resulting scour due to highway encroachments. The effect of lateral channel migration is determined by solving the sediment-continuity equation in the transverse direction. Further adjustment of cross-sectional area is made to reduce the spatial variation in power expenditure along the channel. and grain size. Bank lines are assumed to be stable and fixed in the HEC-6 computation. determines what takes place in the lateral or vertical direction.The channel widening/narrowing is accomplished by coupling a stream tube computer model with a decision-making algorithm using rate of energy dissipation or total stream power minimization. discharge. HEC-6. This uncoupled model was developed at San Diego State University by Chang (1998) to simulate one dimensional. Finally. It is this component that allows the lateral changes in channel geometries. numerical scheme in solving for the two unknown variables of water discharge and cross-sectional area. Army Corps of Engineers (HEC 1993). energy slope. The model is not suitable for rapidly changing flow conditions but can be applied to predict reservoir sedimentation. Sediments are routed by individual size fraction. the fixed-width streamtube computer model. HEC-6 is strictly a one-dimensional model with no provision for simulating the development of meanders or specifying a lateral distribution of sediment transport rate across the section. The HEC-6 program (Scour and Deposition in Rivers and Reservoirs) was developed at the Hydrologic Engineering Center of the U. The second component. flow-continuity and flow-momentum equations in one time step by neglecting storage effects due to unsteady flow. The flow information is then used to compute the bed sediment discharge at each section using one of six user selected sediment transport functions. The quasi-steady backwater equation is used to compute water-flow conditions uncoupled from the sediment-continuity equation.

The success of the numeral model depends on the capabilities of the model. consequently. The exclusion of even one of these three requirements may lead to serious errors in computer simulations. then updates the model geometry and cycles back to RMA-2V to update the flow field. Both finite-element and finite-difference numerical schemes have been applied. During degradation. Even if adequate data are provided for a study river. Included in the first set of data are depths of alluvium and controls such as structures and bedrock outcrops. excluding that portion in the wash load size range. Materials eroded from the channel banks. Examples of finite-element models are the SED2D (Letter et al. If the stream carries a load in excess of its capacity. one can hardly be provided with a complete set of input data in any prototype numerical application. scour and fill. Also under development are Flo2DH Version 3 and Flo1D (Arneson 2001). and stage hydrographs at the upstream and downstream boundaries. However. FESWMS Version 3 will incorporate sediment transport computations. are included in the accounting. including a cross-sectional profile and bed-material size distribution at each computational cross section. resulting in the coarsening of the bed material and formation of an armor coat. and (3) bed-roughness characteristics at each computational point. In most natural rivers. In this sense. quantitative expressions of bed-load and suspended-load discharges. a great number of assumptions often have to be made to fill the gap in the input data. This thickness is a function of the material size and composition. (2) accurate boundary conditions such as water and sediment inflows along the boundaries.47 .Chang (1998) states that the actual sediment rate is obtained by considering sediment material of all size fractions already in the flow as well as the exchange of sediment load with the bed using the method by Borah et al. However. size distributions of boundary-sediment input. It is clear that a computer simulation would be meaningless without the first and second requirements. only extremely limited field data are available for high flood stages at which major riverbed changes occur. Other sources of hydraulic computations can be used as input to SED2D. Thickness of the active layer is based upon the relation defined by Borah. 5. adequate calibration or verification of the models normally cannot be obtained. several of these layers may be scoured away. but also reflects the flow condition. Therefore. et al.3 Data Needs Common to all alluvial river-flow models are requirements for the following input information: (1) accurate initial conditions. new active layers may be deposited on the bed in the process of aggradation.6. During sediment removal. In the case of erosion. Bed armoring develops if bed shear stress is too low to transport any available size. there still remains a need to calibrate and verify the model by means of field data. 5. the quality of the input data and the abilities of the modeler. (1982). These models are best suited for short-term sediment transport simulations although SED2D has features that allow for long-term simulations. it will deposit the excess material on the bed. and. resulting in erroneous feedback of flow information to the riverbed. the capability of the alluvial river-flow models can best be assessed according to how accurately they can predict riverbed changes with limited sources of input data. and the lack of the third requirement would yield an erroneous estimation of flow characteristics. Two-Dimensional Sediment Transport Models are also available. the exchange between the flow and the bed is assumed to take place in the active layer at the surface. SED2D is an uncoupled sediment transport model that generally uses RMA-2V (USACE 1997) hydraulic simulation results to compute sediment transport. 1998) and Flo2DH (Froehlich 1996) models. any size fraction available for entrainment at the bed surface will be removed by the flow and added to the sediment already in transport. A numerical modeler should be aware of which input information is most important to the final result of predicting riverbed changes.

it is often possible to anticipate many impacts on bank stability and provide adequate bank protection in advance. However. but they are unlikely to be far-reaching. lateral erosion problems often occur as a consequence of these changes. There is some evidence that degradation. Construction of a bridge and approach embankments may also have consequences. human activities resulting in gradation problems can be grouped into the following categories: (1) channel alterations. unstable banks and serious debris problems. abutments. their number is so small that a separate category is not warranted. Because human activities dominate the causes for gradation problems. which increases backwater effects upstream of bridges and culverts. 5. however. From an analysis of case histories. Some activities have had far-reaching consequences on streams and have caused. migrate laterally and encroach upon roadways. or stable condition. pile bents.1 Changes Due to Human Activities Human activities are literally changing the face of the Earth and generate accelerated erosion from watersheds. Channel Alterations.7. clearing and snagging. Examples of channel response to straightening are presented in Keefer et al. Because accelerated erosion is associated with human activities. Another very common problem arises when meandering streams. Degradation has also been found to undermine bank protection resulting in the instability of channel banks and increasing debris problems. will be most rapid during a period shortly following the alteration and will thereafter occur at a decreasing rate. Many of the straightened channels have degraded.5. and (4) dams and reservoirs. Some gradation changes should perhaps be classified as being caused by a combination of both natural and human-induced factors. Straightening. (2) land use changes. which undermines abutments and piers located near the bank line. Problems associated with degradation are undermining of footings. (1980). and (2) natural causes or factors. dredging. cutoff walls.7 HIGHWAY PROBLEMS RELATED TO GRADATION CHANGES Gradation problems at highway crossings include changes in the vertical dimension such as aggradation and degradation.48 . Channel straightening is the dominant activity. 5. they will be discussed first. A common problem associated with lateral erosion is bank slumping. The increase in channel slope increases the velocity and the shear stress on the bed. and other alterations of natural channels are the major causes of streambed elevation changes. if it is to occur as a consequence of channel alteration. As a result. or contributed to aggradation and degradation problems at bridges. Human activities were found to be the major cause of streambed elevation changes. Causes of gradation changes that have an impact on highway crossings can be grouped into two basic categories: (1) the result of human activities. The highway problem most associated with aggradation is reduction of flow area. and other flow-control or crossing structures. The degradation is attributed to an increase in channel slope that results from shortening of channel length. artificial constrictions. An analysis of case histories indicates that very few gradation changes were due to natural factors. (3) streambed mining/excavation. the channel bed degrades until the bed becomes armored or the channel widens and begins to meander to reduce the channel slope back to an equilibrium. and degradation is usually accompanied by widening of the channel.

Dams And Reservoirs. The consequences include clear-water releases. channel widening. The response of the system to the increased discharge is to increase channel width and reduce the radius of curvature. Williams and Wolman 1984 and Lagasse 1994). if a sand-bed channel becomes armored with gravel. strip mining. which are then subject to degradation after the mining ceases.49 . Channel widening and reduced sinuosity are common. Streambed Mining/Excavation. In addition. controlled irrigation canal releases. regulated flows. Grazing along the streambanks may have significant effects on bank stability. Equation 5. Natural vegetation is extremely important in maintaining channel stability. are known to increase both discharge and sediment load. The combination of increased peak runoff rates and reduced sediment loads results in channel degradation. has very probably been affected by the clearing of natural vegetation. the amount may be small. backwater. Mining in an upland area may cause aggradation of channels. consider up. and high. The lateral stability of most streams in the United States. Downstream from a reservoir. aggradation may occur downstream from the dam because the flow releases are 5. Downstream mining can also produce headcutting through a bridge waterway. This effect has been documented for many streams (see for example.28 provides a subjective tool for analysis of gradation changes. The total amount of degradation is difficult to predict.Land Use Changes. and unregulated logging are other activities that cause gradation problems. Urban areas are also low sediment producers because of the large percentage of land covered by impervious surfaces. regulated flows. Mining operations upstream of the bridge can also produce degradation at the bridge site and endanger the structure. removal of gravel from pits or trenches in or along the stream may result in a change in flow alignment at a bridge. channel degradation is to be expected because of removal of sediment. Urbanization normally causes significant increase in the magnitude of runoff events while reducing their duration. The response to deforestation and agricultural activities is generally toward increased peak flows and increased sediment yield. undermining the structure. dam breach or removal. as a minimum. Highway engineers should. If sand or gravel is removed from an alluvial channel in quantities that represent a substantial percentage of the annual bedload in transport. on the other hand. agriculture. high sustained. Because this clearing has occurred more or less gradually over the past hundred years.and downstream factors that might cause gradation problems during scheduled bi-annual inspections of bridges. and a reduction in channel sinuosity. Improper construction activities. low sustained. the magnitude of the effect at a particular crossing site is sometimes difficult to assess. The removal of the vegetative cover accelerates the erosion process. On gravel-bed streams. particularly in regions where agriculture or lumbering is practiced. The effects of dams and reservoirs on a stream are complex. the stream will increase its tendency for bank erosion. Urbanization. In response to increased sediment load. the channel will probably degrade.

7. 5. Countermeasures presently used by highway agencies include: (1) sufficient freeboard. Floating Debris.S. Debris hazards occur more frequently in unstable streams where bank erosion is active. (1984) documented a bridge for which a concrete bridge beam was demolished by point impact during a debris flow event. Debris may partially or totally block waterways. meandering/migration (natural cutoffs). debris flows and landslides. 2001) summarizes the results of studies by the U. and (7) routine and emergency removal of debris at bridge crossings. Most debris transported in floods does not travel a great distance and often is observable locally along the streambanks upstream from the bridge prior to the flood. floating debris. Natural causes and complications from gradation problems include: alluvial fans. Debris usually moves as individual logs in a non-random path concentrating in the thalweg of the stream. are associated with most aggrading situations. Mud Flows And Debris Flows. Rapid lowering of river stage may result in severe bank slumping from pore-water pressures in the banks. natural armoring. recurrent flooding.2 Natural Causes Although problems resulting from natural causes are not as frequent as those resulting from human activities. Interbasin transfers of flow and diversions result in periods of channel instability and bank erosion until the new channel regime is established.insufficient to transport gravel brought in by tributary streams. channel bed and bank material erodibility. high stream velocity. both by reduction of flood peaks and a reduction of lateral erosion rates. braiding. in contrast to headwater streams. 5. Many debris problems exist in forested areas with active logging operations. Examples of debris control structures for culverts are given in Reihsen (1964). (3) solid piers. the more general effect of reservoirs is probably to reduce hydraulic problems at highway crossing bridges. create adverse hydraulic conditions that erode pier foundations and bridge abutments or may overtop roadways and cause structural damage. it is important to recognize natural causes in both design and maintenance of highway crossings. threatening bridges and highway structures. Therefore. There is considerable evidence of damages to highway structures in the literature. (2) proper pier spacing. Debris hazards are generally a local phenomena often associated with large floods. (5) special superstructure designs.50 . Channel avulsions. mud and debris flows. and landslides. which can present a serious threat to many engineering structures. (4) debris deflectors. volcanic activity.. Geological Survey to develop methods to estimate potential debris accumulations at bridges (Diehl 1997 and Diehl and Bryan 1993). tectonic activity. The problems are the greatest in the Pacific Northwest and the upper and lower Mississippi River Valley. For example Hungr et al. methods for evaluating its abundance and for mitigating its hazard are deemed feasible. Highway crossings on streams where stream slopes are mild or moderate. However. Most bridge destruction from debris is due to accumulation of debris against bridge components. fire. earthquakes. Floating debris causes hydraulic problems at highway crossings nationwide. (6) flood relief structures. HEC-20 (Lagasse et al. are more vulnerable to debris related hazards. Fast melting snowpack and overabundance of soil moisture on steep slopes throughout the Western United States causes mudflows.

abutments. (1980). base level.8. bed slope. The parameters most influenced by grade changes are those used as input to the hydraulic design procedures currently in use. and flow opening size. energy slope. footing depth at piers and abutments. bridge deck clearance. Examination of typical watershed behavior and response provides information on the impact of changes on the fluvial system. Problem 3. cross-sectional geometry. inundating buildings. Channel stability assessments and possible gradation changes are indicated in Table 5. shear stresses. 5. However. These input parameters include: design discharge. (1980) reviewed design practices to evaluate crossing design procedures and the effect of grade changes on these procedures. Helens triggered a major slope failure on the north flank of the mountain. the volcanic eruption from the magmatic blast of Mount St. Degrading stream reaches affect culvert systems by reducing their structural stability. and construction depth for flow-control and debris-control structures. channel roughness.51 . The upper Cowlitz and the lower Toutle have shifted from meandering to braided streams. The design components most often influenced are culvert capacity and structural stability. The greatest danger produced by aggradation is partial plugging of the culvert opening resulting in a damming effect and increasing the magnitude and frequency of flooding upstream of the structure. 5. and eventually blocking the navigation channel of the Columbia River.9. aggradation produces the most severe problems. backwater. An example relating to the use of this table is presented in Section 5. and flow alignment. Other components of crossing design affected by grade changes include foundation depth. Those findings reflect the observations of Keefer et al. Foundation depths for piers. and the "normal" hydraulic conditions at a site used as input to local scour computations will be changed. The important components of bank protection design adversely affected by grade changes are key depths and the vertical extent of bank protection above and below the streambed. Problems encountered at bridge crossings include bridge capacity.In another example. Bradley (1984) reported the Cowlitz has aggraded markedly as a result of the post eruption hyperconcentrated flows. Mudflows and debris flows were generated and swept down the Toutle and Cowlitz Rivers destroying bridges. velocities. flow depth. debris problems associated with degradation can also have a significant impact on flow capacity and scour.7. With respect to bridge capacity and backwater. General streambed degradation has undermined the foundations of culverts resulting in their complete failure. Problems encountered at culvert crossings can be the result of general grade changes produced by long-term changes in stream morphology or inadequate design and/or construction of culvert systems.3 Resulting Problems at Highway Crossings Brown et al. and flow-control structures can be influenced in two ways by grade changes: the normal streambed base level will be altered. thereby causing difficulties in preventing the failure of some remaining bridges. pier and abutment alignment.

8. However. Observed Channel Response Condition Stable Unstable Degrading Aggrading Alluvial Fan Upstream X X Downstream X X Dam and Reservoir Upstream Downstream River Form Meandering Straight Braided Bank Erosion Vegetated Banks Headcuts Diversion Clear Water Diversion Overloaded with Sediment Channel Straightened Deforested Watershed Drought Period Wet Period Bed Material Size Increase Decrease X X X X X X X X X X X X X X X X X X X X X X X Unknown Unknown Unknown Unknown Unknown X X X Unknown Unknown Unknown Unknown Unknown Unknown 5. a stable channel is one that does not change in size. Channel Response to Changes in Watershed and River Condition (after Keefer et al. an unstable channel is one whose rate or magnitude of change is great enough to be a significant factor in the 5. bridge design.Table 5. all alluvial channels change to some degree and therefore have some degree of instability. and countermeasure placement. For engineering purposes. the annual damage related to hydraulic problems at bridges and highways has been estimated as high as $60 million during years of extreme floods. form. or position through time.8 STREAM STABILITY PROBLEMS AT HIGHWAY CROSSINGS In the United States. Damages by streams can be reduced by considering channel stability in site selection. 1980). Ideally.52 .

53 . Streams whose width at bends is about twice or more the width at straight reaches are called wide-bend streams. 5. The crescentic scars of slumping may be visible from place to place along the bankline. and chute cutoffs with relatively unstable channels. Along an unstable channel. a new meander loop soon forms at the point of cutoff and grows in the same direction as the previous meander.8. Stability is inferred mainly from the nature of point bars. which occurs either by gradual closure of the neck (neck cutoffs) or by a chute that cuts across the neck (chute cutoffs). Neck cutoffs are associated with relatively stable channels.1 Bank Stability On a laterally unstable channel. Cutoffs tend to induce rapid bank erosion at adjacent meander loops. Unstable Banks With Moderate To High Erosion Rate. and (3) natural short-term fluctuations of streambed elevation that are usually associated with the passage of a flood (scour and fill). The presence of a cut bank opposite of a point bar is evidence of instability. Recently formed oxbow lakes along a channel are evidence of recent lateral migration. and a cover of woody vegetation is rarely present. but it may be covered with annual vegetation and low woody vegetation. However. Lagasse et al. Sand or gravel on the bar appears as a light tone on airphotos. At a bend. A recently formed lake is usually immediately adjacent to the channel and it transmits flow at high river stages.planning or maintenance of a highway crossing during the service life of the structure. the point bars are usually wide and unvegetated and the bank opposite to a point bar is cut and often scalloped by erosion. Bars that occur alternately from one side to the other of a straight reach are somewhat analogous to point bars and are indicative of a meandering thalweg. the presence or absence of cut banks. (2) degradation or aggradation of the streambed that continues progressively over a period of years. the point bar opposite of an unstable cut bank is likely to be bare at normal stage. or at actively migrating bends on an otherwise stable channel. and straight reaches tend to be relatively stable. 5. and the variability of stream width (see also HEC-20. Commonly. Fissures. 2001). the channel tends to be wider at bends. However. especially willows. The slope angle of unstable banks usually exceeds 30 percent. even if the point bar is vegetated. Oxbow lakes are formed by the cutoff of meander loops. The kinds of changes considered here are: (1) lateral bank erosion. If the width of an unvegetated point bar is considered as part of the channel width. The presence of abundant oxbow lakes on a floodplain does not necessarily indicate a rapid channel migration rate. Where very rapid erosion is occurring. the bankline may have irregular indentations. such as climate and the timing of floods. The unvegetated condition of the point bar is attributed to a rate of growth that is too rapid for vegetation to become established. meandering of the thalweg in a straight reach is likely to be a precursor of instability. The following paragraphs summarize the characteristics of unstable and stable banks. because an oxbow lake may persist for hundreds of years. the establishment of vegetation on a point bar is dependent on other factors besides rate of growth. For a more detailed discussion of bank stability and the mechanics of bank failure see HEC-20 (Lagasse et al. bank erosion tends to be localized at bends. 2001). which represent the boundaries of actual or potential slump blocks along the bankline indicate the potential for very rapid bank erosion.

Assume that the alignment is fixed by constraints in the acquisition of the right-of-way. In most regions of the United States. Braided streams without point bars (diamond symbol. as are drainage area and discharge. depending on bank height and flow regime of the stream.27c. An example on the use of these results is presented in Section 5. 5. Suppose a meandering river is to be crossed with a highway. When a river crossing site is so constrained by non-hydraulic factors that consideration of alternative sites is not possible. the dashed curve in Figure 5.8.54 . In Figure 5. and what criteria to use to establish the cross-sectional dimensions. For a given channel width.26 channel width is taken as a measure of stream size.26) plot well below the arbitrary curve because their channels are very wide relative to their discharges. Where banks are high. Also. occasional slumps may occur on even the most stable graded banks. Stable Banks With Very Slow Erosion Rate. a change in the river channel alignment is advantageous.27a. and progresses by smoothing of the slope and the establishment of vegetation. dense vegetation may extend to the water's edge at normal stage. but the lower part may be bare at normal stage. Field information on lateral migration rates for channels of different sizes has been compiled by Brice (1982). If a bank is partly graded (smooth slope) the degree of instability is difficult to assess and reliance is placed mainly on vegetation. If braided streams and braided point-bar streams (which are uncommon in most parts of the United States) are excluded. In either case. Figure 5. equiwidth streams tend to have the lowest erosion rates.9. Channel width is an imperfect measure of stream size. The grading of a bank typically begins with the accumulation of slumped material at the base such that a slope is formed. and braided point-bar streams the highest. as shown in Figure 5. consideration for improvement to the channel would also be advisable for a hypothetical lateral encroachment of a highway as depicted in Figure 5.26 provides a preliminary estimate of erosion rates that may be encountered at a particular site. The dashed line is drawn arbitrarily to have a slope of 1 and a position (intercept) to separate most equiwidth streams from most wide-bend and braided point-bar streams. consideration is given to channel improvement as shown in Figure 5. To create better flow alignment with the bridge. Bank erosion rates tend to increase with increasing stream size. Where banks are low. particularly for the comparison of streams in arid and semiarid regions with streams in humid regions. Stable banks tend to be graded to a smooth slope and the slope angle is usually less than about 30 percent. the engineer must attempt to improve the local situation to meet specific needs. Mature trees on a graded bank slope are particularly convincing evidence for bank stability.Unstable Banks With Slow To Moderate Erosion Rate. Similarly. Problem 4. the upper parts of stable banks are vegetated. the engineer may be forced to make channel improvements in order to maintain and protect existing highway structures in or adjacent to the river. 5.27b. Shallow mountain streams that transport coarse bed sediment tend to have stable banks. the designer's questions are how to realign the channel.2 Stability Problems Associated With Channel Relocation For some highway encroachments.

5.55 .26. Median bank erosion rate in relation to channel width for different types of streams (after Brice 1982).Figure 5.

27) are established by existing boundary conditions.Figure 5. When the angle φ defined in Figure 5. Prior to realigning a river channel the stability of the existing channel must be examined using the methods outlined earlier. it is important for the designer to understand and appreciate river hydraulics and geomorphology when making decisions concerning channel relocation. (designated 1 and 2.22) 5.15 exceeds about 40 degrees. it is necessary to stabilize the outside banks of the curves in order to hold the new alignment. the total drop in bed elevation for the new channel (subscript 2) and the old channel (subscript 1) are the same. ∆z1 = ∆z2 = ∆z (5. If curves are included. As indicated in the previous chapters. The sinuosity and channel bed slope are related in the following way. Different rivers have different characteristics and historical background with regard to channel migration. it is possible to provide some principles and guidelines for the design engineer. this provides a sufficient crossing length for the thalweg to shift from one side of the channel to the other. The bed elevations at the ends of the reach being rechannelized. Nevertheless. Knowledge about river systems has not yet advanced to such a state as to make this possible. and depending upon crossing length some amount of maintenance may be necessary to remove sandbars after large floods so that the channel does not develop new meander patterns in the crossings during normal flows. Hence.56 . The realigned channel may be made straight without curves. Generally. As the general rule. Encroachment on a meandering river. discharge. the radii of bends should be made about equal to the mean radii of bends in extended reaches of the river. recent and past aerial photographs and field surveys are generally necessary. the radii of curvature. stage. in Figure 5. It is difficult to state generalized criteria for channel relocation applicable to every river. or may include one or more curves. the number of bends. the limits of rechannelization (hence the length or slope of the channel) and the cross-sectional area are decisions which have to be made by the designer. A stream classification. geometry and sediment transport.27.

The length of channel measured along the thalweg is labeled Lc. 5.27) but L V1 = L V2 = L v (5. Thus. As indicated before.26) Sn2 = L c2 Lv 2 (5. hence Sn2. The new channel alignment. The new slope S2 can be calculated from Equation 5. the mean slope of the channel bed before relocation is: S1 = ∆z L c1 (5.1) should be satisfied. If Sn2 < Sn1 then S2 > S1. deflection angles and tangent lengths between reversing curves. consideration should also be given to prevailing average conditions in the extended reach.30. æ L c ö æ ∆z ö æ L c 2 ö æ ∆z ö Sn1S1 = ç 1 ÷ ç ÷ ç ÷ =ç ÷ = s n 2 S2 è L v ø è L c1 ø è L v ø è L c 2 ø (5.57 .23) and after relocation is S2 = ∆z L c2 (5. and the relationship for meandering (Equation 5.30) The new channel slope and channel sinuosity are inversely related. Lc to the length of the valley. can be chosen by the designer with due consideration given to the radii of curvature.24) Sinuosity is defined as the ratio of the length of channel.29) Thus.28) and ∆z1 ∆z2 ∆z = = L v1 L v2 Lv (5.25) Clearly. or Sn = Lc ≥1 Lv (5. Sn1 = L c1 Lv 1 (5.

S 2 Q1/ 4 ≤ .0007

(SI)

or

S2 Q1/ 4 ≤ 0.0017 (English)

If S1 is of such magnitude that Equation 5.1 cannot be satisfied with still larger S2, the possibility of the river changing to a braided channel because of steeper slope should be carefully evaluated. With steeper slope, there could be an increase in sediment transport which could cause degradation, and the effect would be extended both upstream and downstream of the relocated reach. The meander patterns could change. Considerable bank protection might be necessary to contain lateral migration which is characteristic of a braided channel, and if the slope is too steep, head cuts could develop which migrate upstream with attendant effects on the geometry of the channel. Even when changes in slope are not very large, a short-term adjustment of the average river slope occurs, consistent with the sediment transport rate, flow velocities and roughnesses, beyond the upstream and downstream limits of channel improvement. For small changes in slope, the proportionality (Equation 5.16), QS ~ Qs D50 tends toward equilibrium with slight increases in bed sediment size, D50, and adjustment in the sediment transport rate, Qs. A small increase in the new channel width could be considered which tends to maintain the same stream power, τo V, in the old and new channels. That is,

( τ o V )1 = ( τ o V ) 2

(5.31)

**With substitution of τo = γRS, V = Q/A and R = A/P
**

W2 = S2 W1 / S1

_ A/W, Equation 5.42 leads to

(5.32)

Any designed increase in width should be limited to about 10 to 15 percent. Wider channels would be ineffective. Deposition would occur along one bank and the effort of extra excavation would be wasted. Furthermore, bar formation would be encouraged, with resultant tendencies for changes in the meander pattern leading to greater maintenance costs for bank stabilization and removal of the bars to hold the desired river alignment. The depth of flow in the channel is dependent on discharge, effective channel width, sediment transport rate (because it affects bed form and channel roughness) and channel slope. Methods for evaluating flow depth were discussed earlier in this chapter. The foregoing discussion pertains to alluvial channels with silt and sand sized bed materials. For streams with gravel and cobble beds, the usual concern is to provide adequate channel cross-sectional dimensions to convey flood flows. If the realigned channels are made too steep, there is an increased stream power with a consequent increase in transport rate of the bed material. The deposition of material in the reaches downstream of the crossing tends to form gravel bars and encourages changes in the planform of the channel. Short-term changes in channel slope can be expected until equilibrium is reestablished over extended reaches both upstream and downstream of the rechannelized reach. Bank stabilization may be necessary to prevent lateral migration, and periodic removal of gravel bars may also be necessary. 5.58

5.8.3 Assessment of Stability for Relocated Streams Brice (1980) reported case histories for channel stability of relocated streams in different regions of the United States. Based on his study, the recommendations and conclusions presented here apply to specific aspects of the planning and construction of channel relocation. They are intended for assessment of the risk of instability and for reduction of the degree of instability connected with relocation. Serious instability resulting from relocation can be observed either when the prior natural channel is unstable or when floods of high recurrence interval occur during or soon after construction. Although there is an element of uncertainty in channel stability, the experience represented by Brice's study sites provides useful guidelines for improvement in the performance of channels relocated by highway agencies. Consideration of the following aspects of the channel relocation is recommended. Channel Stability Prior To Relocation. Assessment of the stability of a channel prior to relocation is needed to assess erosion-control measures and risk of instability. An unstable channel is likely to respond unfavorably to relocation. Bank stability is assessed by field study and the stereoscopic examination of aerial photographs. The most useful indicators of bank instability are cut or slumped banks, fallen trees along the bankline, and wide, unvegetated, exposed point bars. Bank recession rates are measured by comparison of time-sequential aerial photographs. Vertical instability is equally important but more difficult to determine. It is indicated by changes in channel elevation at bridges and gaging stations. Serious degradation is usually accompanied by generally cut or slumped banks along a channel. Erosion Resistance Of Channel Boundary Materials. The stability of a channel, whether natural or relocated, is partly determined by the erosion resistance of materials that form the wetted perimeter of the channel. Resistant bedrock outcrops, which extend out into the channel bottom, or that lie at shallow depths, will provide protection against degradation. Not all bedrock is resistant. Erosion of shale, or of other sedimentary rock types interbedded with shale, has been observed. Degradation was slight or undetected at most sites where bed sediment was of cobble and boulder size. However, serious degradation may result from relocation. Degradation may result from the relocation of any alluvial channel, whatever the size of bed material, but the incidence of serious degradation of channels relocated by highway agencies is slight. The cohesion and erosion resistance of banks tend to increase with clay content. Banks of weakly coherent sand or silt are clearly subject to rapid erosion, unless protected with vegetation. No consistent relation was found between channel stability and the cohesion of bank materials, probably because of the effects of vegetation. Length Of Relocation. The length of relocation contributes significantly to channel instability at sites where its value exceeded 250 channel widths. When the value is below 100 channel widths, the effects of length of relocation are dominated by other factors. The probability of local bank erosion at some point along a channel increases with the length of the channel. The importance of vegetation, both in appearance and in erosion control, would seem to justify a serious and possibly sustained effort to establish it as soon as possible on the graded banks. Bank Revetment. Revetment makes a critical contribution to stability at many sites where it is placed at bends and along roadway embankments. Rock riprap is by far the most commonly used and effective revetment. Concrete slope paving is prone to failure. Articulated concrete block can be effective especially when vegetation can establish in the interstices between blocks. Bank revetment is discussed in detail in the next chapter.

5.59

Check Dams (drop structures). In general, check dams are effective in preventing channel degradation. The potential for erosion at a check dam depends on its design and construction, its height and the use of revetment on adjoining banks. A series of low check dams, less than about 0.5 m (2 ft) in height, is probably preferable to a single higher structure, because the potential for erosion and failure is reduced. By simulating rapids, low check dams may add visual interest to the flow in a channel. One critical problem arising with check dams relates to improper design for large flows. Higher flows have worked around the ends of many installations to produce failure. Maintenance. The following problems that can be controlled by maintenance were observed along relocated channels: (1) growth of annual vegetation in channel; (2) reduction of channel conveyance by overhanging trees; (3) local bank cutting; and (4) bank slumping. The expense of routine maintenance or inspection of relocated channels beyond the highway right-of-way is probably prohibitive. However, most of the serious problems could be detected by periodic inspection, perhaps by aerial photography, during the first 5 or 10 years after construction. Relationship Between Sinuosity And Stability. This relationship is summarized as follows: (1) Meandering does not necessarily indicate instability; an unstable stream will not remain highly sinuous for very long, because the sinuosity will be reduced by frequent meander cutoffs; (2) Where instability is present along a reach, it occurs mainly at bends; straight segments may remain stable for decades; and (3) The highest instability is for reaches whose sinuosity is in the range of 1.2 to 2 and whose type is either wide bend or braided point bar. 5.8.4 Estimation of Future Channel Stability and Behavior One objective of stability assessment is to anticipate the migration of bends and the development of new bends. Lateral erosion is probably more frequently involved in hydraulic problems at bridges than any other stream process. Problems caused by general scour, local scour, channel degradation and accumulation of debris are somewhat less common. The lateral stability of a stream can be measured from records of its position at two or more different times where the available records are usually maps or aerial photographs. Historic surveyed cross-sections are extremely useful and can frequently be located in bridge inspection files. It is recognized that some progress is being made on the numerical prediction of loop deformation and bend migration (see Section 5.8.5). At present, however, the best available estimates are based on past rates of lateral migration at a particular reach. However, erosion rates may fluctuate substantially from one period of years to the next. Measurements of bank erosion on two time-sequential aerial photographs (or maps) require the identification of reference points which are common to both. Useful reference points include roads, buildings, irrigation canals, bridges and fence corners. This analysis of lateral stability is greatly facilitated by a drawing of time changes in bankline position. To prepare such a drawing, aerial photographs are matched in scale and the photographs are superimposed holding the reference points fixed. For further discussion see HEC-20 (Lagasse et al. 2001). Bank erosion rates increase with the stream size as shown in Figure 5.26. Sinuous canaliform streams can then be expected to have the lowest erosion rates and the sinuous braided streams, the highest.

5.60

The lateral stability of different stream reaches can be compared by means of a dimensionless erosion index. The erosion index is the product of its median bank erosion rate expressed in channel widths per year, multiplied by the percent of reach along which erosion occurred, multiplied by 1,000. Erosion indexes for 41 streams in the United States are plotted against sinuosity in Figure 5.28. The length of most of these reaches is 25 to 100 times the channel width. The highest erosion index values are for reaches with sinuosity ranging between 1.2 and 2. Erosion indexes are large for sinuous braided and sinuous point bar (wide bend) streams. Equiwidth streams tended to be relatively stable. The erosion index value of 5, in Figure 5.28, is suggested as a boundary between stable and unstable reaches. Brice (1984) considers that reaches having erosion indexes values less than 5 are unlikely to cause lateral erosion problems at bridges. An example on the use of Figure 5.28 is presented in Section 5.9, Problem 4. A general assessment of bank stability can be made considering the following aspects. Bank Erosion Rates. It is theoretically possible to determine bank erosion rates from factors such as water velocity and resistance of the banks to erosion. See HEC-18 (Richardson and Davis 2001) for a discussion of recent advances in measuring erosion rates. The results in Figure 5.26 provide a first approximation of migration rate of a bend regardless of the hydraulic conditions and sediment characteristics. Past rates of erosion at a particular site provide the best estimate of future rates. In projecting past rates into the future, consideration must be given to the following factors: (1) the past flow history of the site during the period of measurement, in comparison with the probable future flow history during the life span of the highway crossing. The duration of floods, or of flows near bankfull stages, is probably more important than the magnitude of floods; and (2) human-induced factors that are likely to affect bank erosion rates. Among the most important of these are urbanization and the clearing of floodplain forests.

Figure 5.28. Erosion index in relation to sinuosity (Brice 1984).

5.61

Behavior Of Meander Loops. If the proposed bridge or roadway is located near a meander loop, it is useful to have some insight into the probable way in which the loop will migrate or develop, as well as its rate of growth. No two meanders will behave in exactly the same way, but the meanders on a particular stream reach tend to conform to one of the several modes of behavior illustrated in Figure 5.29. Mode A (Figure 5.29) represents the typical development of a loop of low amplitude, which decreases in radius as it extends slightly in a downstream direction. Mode B rarely occurs unless meanders are confined by valley sides on a narrow floodplain, or are confined by artificial levees. Well developed meanders on streams that have moderately unstable banks are likely to follow Mode C. Mode D applies mainly to large loops on meandering or highly meandering streams. The meander has become too large in relation to stream size and flow, and secondary meanders develop along it, converting it to a compound loop. Mode E also applies to meandering or highly meandering streams, usually of the equiwidth point-bar type. The banks have been sufficiently stable for an elongated loop to form (without being cut off), but the neck of the loop is gradually being closed and cutoff will eventually occur at the neck. Modes F and G apply mainly to locally braided sinuous or meandering streams having unstable banks. Loops are cut off by chutes that break diagonally or directly across the neck. Effects Of Meander Cutoff. If cutoffs seem imminent at any meanders in the vicinity of a proposed bridge crossing the probable effects of cutoff need to be considered. The local increase in channel slope due to cutoff usually results in an increase in the growth rate of adjoining meanders, and an increase in channel width at the point of cutoff. On a typical wide-bend point-bar stream the effects of cutoff do not extend very far upstream or downstream.

Figure 5.29. Modes of meander loop development. A, Extension. B, Translation. C, Rotation. D, Conversion to a compound loop. E, Neck cutoff by closure. F, Diagonal cutoff by chute. G, Neck cutoff by chute.

5.62

Assessment Of Degradation. Field sites having degradation problems are more numerous than sites having aggradation problems. Annual rates of degradation averaged from past records such as the closure of a dam give poor estimates of future rates of degradation. Typical situations exhibit an exponential decay function of the rate of channel degradation. Recent evidence of degradation can be detected from field surveys or by stereo viewing of aerial photographs. Indicators of degradation are: (1) channel scarps, headcuts and nickpoints; (2) gullying of minor side tributaries; (3) high and steep unvegetated banks; (4) measurements of streambed elevation from a bridge deck; (5) changes in stream discharge relationships; and (6) measurements of longitudinal profiles. Assessment Of Scour And Fill. Natural scour and fill refer to fluctuations of streambed elevation about an equilibrium condition. These fluctuations are associated mainly with floods and occur by three different mechanisms operating jointly or independently: (1) bed form migration; (2) convergence and divergence of flow; and (3) lateral shift of thalweg or braids. The maximum scour induced by the migration of a dune is almost one-half the dune height, and dune heights are roughly estimated as one-third of the mean flow depth. In gravel bed streams, most migrating bed forms can be regarded as bars, the height of which is related to flow depth. The migration of a bar through a bridge waterway is mainly of concern because of the deflection and concentration of flow. Bar migration tends to be a random process and its motion can best be tracked from time-sequential aerial photographs. Gravel bars tend to migrate on braided streams and to remain fixed at riffles on unbraided pool and riffle streams. Flow convergence in natural streams is associated with scour, whereas divergent currents are associated with deposition. Persistent pools have the strongest convergence of flow and the greatest potential for scour. Such pools are best identified by a continuous bed profile along the thalweg. In braided streams, scour holes are found at the confluence of braids. Field measurement of cross-sectional area and flow velocity at an incised reach near bankfull stage provides a good basis for calculation of scour by extrapolation to the design flood. Instability of the streambed that results from shift of thalweg is related to stream type and can be assessed from study of aerial photographs. On sinuous canal form streams, shift of the thalweg during flood is minimal. A greater shift of the thalweg can be expected on sinuous point-bar streams. In straight reaches, alternate bars visible on aerial photographs taken at low stage are commonly present. These alternate bars indicate the potential for thalweg shifting and also for bank erosion when the current is deflected against the bankline. Shift of the thalweg with increase in stage must be considered when determining the location of the point of maximum bed scour, bank erosion, and the alignment of piers with flood flow. Site Selection For Highway Crossings. For most streams the magnitude of scour is substantially greater at specific places along the channel. Bends and narrow sections may scour at high stages regardless of the effect of bridge structures. Straight or gently curved reaches with stable banks are preferred. Considerations for the selection of a crossing site on a non-sinuous reach include: (1) is the site at a pool, riffle or transition section; (2) are alternate bars visible at low stage; and (3) what is the effect of migration of mid-channel bars, if any? With respect to meandering reaches, questions requiring solution include: (1) what has been the rate and mode of migration of the meander; (2) what is the probable future behavior, as based on the past; (3) is the site at a pool, riffle or transition section; and (4) is meander cutoff probable?

5.63

5.8.5 Advances in Predicting Meander Migration Meander Behavior. According to Hooke (1991), it has long been assumed that, after initial development, meanders tend towards an equilibrium of form, given stability of external conditions. These ideas were based on early experiments in flumes (Friedkin 1945) and on the analysis of process-form relations in the 1960s in which statistical relationships were established, e.g., between meander wavelengths and discharge (Carlston 1965). Much evidence has now been accumulated that many meanders exhibit no such equilibrium (Carson and Lapointe 1983, Hooke 1984). If the behavior of individual bends on active rivers is examined (using for example, historical evidence or meander scrolls) then a continuous evolution with an increasing complexity of pattern is frequently found. Meanders may migrate at first, but then they begin to grow in amplitude, then to become compound in form. Based on analysis of historical sequences of meander change on rivers using maps and aerial photographs, models of meander development have been produced (Hooke 1991; Harvey 1989; Keller 1972). These show a sequence from low sinuosity bends to bends of symmetrical form which tend to migrate; then bends enter a phase of rapid growth with maximum erosion at the apex leading to extension in the cross-valley direction. Beyond a certain increase in path length (Hooke and Harvey 1983) the bends start to become compound in form by development of lobes on one or both parts of the apex. These lobes may go on to develop into separate bends if there is space. This increasing asymmetry and complexity is similar to other models such as those of Brice (1974) and Hickin and Nanson (1975). This process of growth cannot, of course, go on indefinitely. Eventually the bends are likely to intersect or spatial limits of the floodplain are reached so that cut-offs take place. Therefore, following a phase of growth, cutoffs are likely. Rivers show an early phase of increasing sinuosity then oscillating sinuosity thereafter. Of course, not all bends exhibit complexity or rapid growth. Some simply progressively migrate or tend to stabilize at various stages. This could be for a number of reasons, e.g., slope and overall energy in the system, floodplain form, or it could simply be that changes are progressing at a very slow rate. Different domains of meander behavior can therefore be visualized (Figure 5.30). Theoretically, it should be possible to identify domains of behavior with thresholds between them. The thresholds will vary with the actual river and may be rather 'fuzzy.' Further work is needed to substantiate whether they do indeed exist, but the indication is that any single mathematical model of meander behavior will be inadequate. This is supported by problems with existing models. The mathematical modeling of helical and cross channel flows and their related channel characteristics has proved to be difficult. Most such models are based on simplifications of the equations of continuity for water and sediment and on the Reynold's (or St. Venant's) equations of motion (Engelund 1974; Smith and McLean 1984; Odgaard 1986). As noted by several researchers (Odgaard and Bergs 1987; Yen and Ho 1990), most of these models simulate bend flow and channel topography only in the "fully developed" portions of the bend, i.e., where velocity and thalweg depth do not vary longitudinally. Other models predict these characteristics throughout the bend, even when velocity and depth change downstream (Dietrich and Smith 1983; Engelund 1974; Odgaard 1986; Yen and Ho 1990). In addition, considerable disagreement exists over the simplifications and eliminations that have been used to solve these equations (Dietrich and Smith 1983).

5.64

Figure 5.30. Domains of meander behavior (from Hooke 1991). Most of the mathematical models have been tested or verified in flumes, some of which had fixed beds or constant radius of curvature segments connected to straight segments (Odgaard and Bergs 1987). Even when model testing is done on rivers with constantlyvarying radii of curvature and mobile bed materials, problems with measurement accuracy of the various model components often dominate the final result (Dietrich and Smith 1983). One difficulty is the assumption that all meanders are simple, whereas as Hooke (1991) demonstrates they are not. In addition to other controls on meander migration, the mode of bank retreat varies from grain by grain removal to mass failure (Hasegawa 1989; Knighton 1984). There are many authors who believe that it is combinations of processes that are important (e.g., Hooke 1979; Thorne 1982; Lawler 1986). In view of the wide range of alluvial materials, riverine forms and hydroclimatic environments, the limitations of mathematical models are many and the range of validity of the models is limited. Therefore a reasonable forecast of the future planform of the river has not yet been realized successfully and remains a challenging topic (Thein 1994). Mosselman (1995) states that the non-periodicity of meanders, which develop from a straight alignment in the numerical models of Howard and Knutson (1984) and others "suggest chaotic behavior, that is, a sensitive dependence on initial and boundary conditions." He concludes that this "chaotic behavior bears on the predictability of river meandering." It is only by considering the great variety of meander configurations and how they influence future patterns that reasonable predictions of meander shift can be accomplished. Indeed, Howard (1996) concludes that the chaotic behavior suggests that "even the most detailed model has limited predictive power." This is perhaps because "small differences in initial geometry or boundary conditions between two identical streams will cause different meander patterns" (Howard 1996). In addition, he concludes that "predictability of future meander patterns decrease with time."

5.65

The problem appears to lie in the lack of recognition of different types of rivers with different meander behavior (Figure 5.30) and the assumption that all meanders are represented by idealized bends. Analysis Options. A study by Johns Hopkins University (Cherry et al. 1996) for the U.S. Army Corps of Engineers Waterways Experiment Station investigated the use of both empirical and analytical approaches to provide solutions to the problem of predicting meander migration. This study evaluated empirical and analytical (computational) methods for forecasting planform change and bankline migration in flood control channels, using data originally assembled and analyzed by Brice (1982) to assess stream channel instability problems at bridges for the FHWA. Twenty-six sites were used to evaluate the predictive capabilities of bend-flow meander migration computer models. The computational bend-flow meander migration model used by Cherry et al. was developed by Garcia et al. (1994). The Johns Hopkins study recognized that a meandering river is a complex system involving relations among many variables. The erosion rate for a meander bend is determined by the balance between the erosive forces applied to the channel bank and the resistance to erosion provided by the bank material and bank vegetation. Erosive force is a complex function of discharge, channel cross section geometry, sediment load, bed roughness, presence of bedforms and bars, and the planform geometry. Resistance to erosion is related to the properties of the bank material, the bank geometry (slope, height, shape), the presence of vegetation, and the state of the pore water in the bank (Cherry et al. 1996). Although simplified, single valued correlations between a number of variables were established empirically and expressed as power functions, Johns Hopkins concluded that they did not adequately describe meander behavior. In regard to computer modeling, a number of authors have developed versions of the bend flow model (e.g., Parker et al. 1982; Beck and Melfi 1984; Hasegawa 1989; Odgaard and Bergs 1987; Garcia et al. 1994) and although the models have typically been tested with plots of predicted vs. observed channel form for a limited number of channels, there has been little general testing of these models over a range of hydrologic and geologic conditions (Cherry et al. 1996). After testing the bend flow model for 26 of the meandering sites in the Brice data set, the Johns Hopkins study concluded that both the accuracy and applicability of the bend-flow meander migration model are limited by a number of simplifying assumptions. Among the most important of these are the use of a single discharge and the assumption of constant channel width, both of which prevent the model from successfully forecasting the spatial and temporal variability that appears to be inherent in the process of bend migration. It was also concluded that much of the discrepancy between the predicted and observed distributions of erosion can be accounted for by the fact that meander migration is modeled as a smooth, continuous process. In reality, erosion occurs predominantly in discrete events, and varies greatly both temporally and spatially along the channel from bend to bend (Nanson and Hickin 1983). The Johns Hopkins study noted that the identification of local factors that influence the amount of bank erosion that occurs is a subject "that will require further investigation."

5.66

In addition to channel and bank characteristics, floodplain characteristics must also be incorporated into an analysis procedure. The floodplain characteristics that should affect meander migration include geologic controls, alluvial deposits and topographic variability. Geologic controls include bedrock outcrops and erosion resistant features along the valley sides. Alluvial deposits frequently include oxbows, meander scrolls and scars, and clay plugs, each with different erodibility characteristics. Topographic variability that should be considered include the cross valley slope of the adjacent floodplain and valley slope. These factors could be incorporated into regression equations but would be difficult to include in computer modeling of bendway migration. 5.9 SOLVED PROBLEMS RIVER MORPHOLOGY AND RESPONSE 5.9.1 PROBLEM 1 Meandering and Braiding (a) Consider the sinuous point bar stream in Figure 5.31. Determine the following characteristics: meander wavelength λ; meander width W m; mean radius of curvature rc; meander amplitude A; and the bend deflection angle φ.

Figure 5.31. Sinuous point bar stream. Meander wavelength λ is approximately 11,000 ft (3,353 m) Radius of curvature rc is approximately 2,400 ft (732 m) Channel width ranges from 250 ft to 850 ft (76 to 260 m) at high stage Meander amplitude A is 2,200 ft (670 m) Meander width W m is 2,700 ft (823 m) Bend deflection angle φ is 105° Sinuosity is 1.2 (b) Given the sand size D50 = 0.5mm, the bankfull discharge Q = 10,000 cfs (283 m /s) and -4 the slope S = 2x10 , determine the effect of increasing slope, discharge, sediment size and sediment discharge on the planform geometry.

3

5.67

The result of increasing any or several of these variables (Figure 5.18 and Table 5.4) is to promote braiding of this channel. When locating this stream on Lane's diagram (Figure 5.18) it 0.25 0.25 is shown that with SQ = 0.002 (SI = .00082), this river is very close to the line SQ = 0.0017 (.0007 for SI) for meandering streams. Hence an increase in discharge or slope would be required to change the planform to a braided stream. (c) Determine the effect of increased discharge of bed sediment size, bed sediment load and washload on channel stability, resistance to flow, energy slope and stage of the same river. The Table 5.4 can be used to provide a qualitative response to these changes. • • • • Increase in discharge results in an increase in stage and a decrease in energy slope and channel stability. Increase in sediment size results in an increase in stage, energy slope and resistance to flow. The channel stability might not be changed. Increase in bed sediment load should increase the channel stability through a decreased resistance to flow, slope and stage. Effect of increasing washload is similar to that of increasing bed sediment load except for channel stability, which is uncertain.

5.9.2 PROBLEM 2 Classification of Alluvial Reaches Identify the three types of alluvial river reaches sketched below. Discuss the relative stability of each channel.

Figure 5.32. River channels (After Petersen 1986).

5.68

Based on the classification presented in Figures 5.11, 5.12, and 5.13, the first stream (a) is a straight channel with alternate bars and sinuous thalweg. This stream has a relatively low slope and low width-depth ratio. Figure 5.13 also indicates that it has an intermediate sediment size, sediment load and moderate ratio of bedload to total load (between 3 - 10 percent). This stream can be classified as relatively stable. The second stream (b) is a sinuous point bar stream which is somewhat wider at bends. The meander bends are expected to shift gradually with possible neck cutoff. The stream is then relatively stable. The third stream (c) is a braided sand-bed stream with multiple bars and channels. Bars are likely to be comprised of coarse sand and the large flow velocities and stream power will generate large sediment load with a large proportion transported as contact load. The overall stability of this braided channel is very low, channel shifting, and avulsions are certainly common. 5.9.3 PROBLEM 3 Channel Response to Changes in Watershed Conditions Determine the effect of watershed deforestation, bank erosion and headcutting on channel stability and gradation changes in an alluvial stream. Referring to Table 5.8, deforestation generally causes aggradation problems and therefore channel instability. The reason for this is that deforestation increases runoff and peak runoff discharge as well as sediment transport from upland areas. Headcutting is a degradation process. Upstream migration of headcuts induces bank failure and channel stability problems. Bank erosion has basically the same consequence on channel stability as that of headcutting. The gradation changes, however, are more difficult to assess because bank erosion changes the width-depth ratio. 5.9.4 PROBLEM 4 Channel Migration Rate (a) Determine the bank erosion rate and the erosion index of the sinuous point bar stream sketched in Figure 5.33. The best method to estimate the rate of bank erosion is to compare two sets of aerial photographs. However, a first assessment can be obtained from Figure 5.26. Entering the figure with an average width around 150 m (492 ft), the median erosion rate should be around 2 meters (6.5 ft) per year. Wide bend streams have slightly larger erosion rates than given by the dashed line. Entering Figure 5.28 with a sinuosity of 1.2, the erosion index might be as large as 18 indicating channel instability. Erosion is to be expected on the concave side of the bends. Other means for assessing lateral migration rates such as study of past aerial photographs, scroll formation (Figure 5.12(b)) and field studies should be undertaken if any bridge crossing was to be built across this river. (b) Sketch the likely future changes in the meandering river shown in the sketch below (Figure 5.34).

5.69

Figure 5.33. Meandering river sketch (after Petersen 1986).

Figure 5.34. Estimated future location of a meandering river (after Petersen 1986).

5.70

The river (left bank) has moved downvalley approximately 6,000 feet (1,829 m) in 50 years or an average of 120 ft (36 m) per year. The narrowest point of the meander neck is about 2,500 ft (762 m). At its historical rate, this meandering channel will most likely lead to a cutoff in the next 15 to 20 years. The estimated position of the channel is sketched in Figure 5.34. 5.9.5 PROBLEM 5 At-A-Station and Downstream Hydraulic Geometry Relationships (SI) At bankfull discharge conditions Q1 = 227 m /s and the width of a sand-bed stream (Ds1 = 0.6 -4 mm) is W 1 = 76 m, the maximum flow depth is yo1 = 2.4 m, the slope is Sf1 = 2.5 x 10 , and the maximum velocity is V1 = 1.5 m/s. (a) Estimate the width, W 2, depth yo2, slope Sf2 and velocity V2 at the same station when the 3 discharge Q2 is 5.7 m /s if the cross-sectional geometry is unknown. The at-a-station hydraulic geometry relationships (Table 5.3) can be used when no specific field data is available. For width:

æ Q2 ö W2 = W1 ç ÷ çQ ÷ è 1ø

0.26

3

æ 5 .7 ö = 76 ç ÷ è 227 ø

0.26

= 29.2 m

For depth:

y o2 æ Q2 ö = y o1 ç ÷ çQ ÷ è 1ø

0.40

æ 5 .7 ö = 2.4 ç ÷ è 227 ø

0.40

= 0.55 m

Slope is unchanged:

S f1 = S f2 = 2.5 x 10 −4

For velocity:

æ Q2 ö V2 = V1 ç ÷ çQ ÷ è 1ø

0.34

æ 5.7 ö = 1.5 ç ÷ è 227 ø

0.34

= 0.43 m / s

(b) Using the same station as for part (a), estimate the width, W 2. depth yo2, slope Sf2 and 3 velocity V2 in an upstream section of this stream if the bankfull discharge is 14 m /s and the bed material is gravel (D50 = 8 mm). How would the hydraulic geometry change if the bed material upstream is sand (D50 = 0.6 mm)?

5.71

The "downstream" geometry relationships can be used in this case. Two types of relationships are given in Table 5.3. The sand bed relationships are a function of discharge only, whereas the gravel bed relationships are a function of both discharge and sediment size. Both methods are compared in the following. Flow Depth Sand bed:

y o2

æ Q b2 ö = y o1 ç ÷ çQ ÷ è b1 ø

0.46

æ 14 ö = 2 .4 ç ÷ è 227 ø

0.46

= 0.67 m

0.4

Gravel bed: Channel width Sand bed:

y o2

æ Q b2 ö = y o1 ç ÷ çQ ÷ è b1 ø

0.40

æ 14 ö = 2.4 ç ÷ è 227 ø

= 0.79 m

æ Q b2 ö W2 = W1 ç ÷ çQ ÷ è b1 ø

0.46

æ 14 ö = 76 ç ÷ è 227 ø

æ D s2 ç ç Ds è 1 ö ÷ ÷ ø

−0.33

0.46

= 21.1 m

Gravel bed:

æ Q b2 ö W2 = W1 ç ÷ çQ ÷ è b1 ø

0.53

æ 14 ö = 76 ç ÷ è 227 ø

0.53

æ 8 ö ç ÷ è 0.6 ø

−0.33

= 7.4 m

**Friction slope Sand bed:
**

S f2

æ Qb2 ö = S f1 ç ÷ çQ ÷ è b1 ø

−0.46

= 2.5 x 10

−0.4

−4

æ 14 ö ç ÷ è 227 ø

−0.46

= 9 x 10 − 4

Gravel bed:

S f2

æ Q b2 ö = S f1 ç ÷ çQ ÷ è b1 ø

æ D s2 ç ç Ds è 1

− 0.4 ö 14 ö æ 8 ö ÷ = 2.5 x 10 − 4 æ ç ÷ ç ÷ = 0.01 ÷ è 0.6 ø è 227 ø ø

1

**Velocity Sand bed:
**

æ Q b2 ö V2 = V1 ç ÷ çQ ÷ è b1 ø

0.08

æ 14 ö = 1 .5 ç ÷ è 227 ø

æ D s2 ç ç Ds è 1 ö ÷ ÷ ø

.33

0.08

= 1.2 m / s

Gravel bed:

æ Q b2 ö V2 = V1 ç ÷ çQ ÷ è b1 ø

0.07

æ 14 ö = 1 .5 ç ÷ è 227 ø

0.07

æ 8 ö ç ÷ è 0.6 ø

.33

= 2 .9 m / s

The gravel bed relationships give a steeper, faster flowing and narrower channel as compared to the sand bed relationships. Unless the sediment size is markedly different for two streams, the resulting hydraulic geometry calculated from both sets of equations will be similar.

5.72

5.9.6 PROBLEM 6 At-A-Station and Downstream Hydraulic Geometry Relationships (English) At bankfull discharge conditions Q1 = 8000 cfs and the width of a sand-bed stream (Ds1 = 0.6 -4 mm) is W 1 = 250 ft, the maximum flow depth is yo1 = 8 ft m, the slope is Sf1 = 2.5 x 10 , and the maximum velocity is V1 = 5 ft/s. (a) Estimate the width, W 2, depth yo2, slope Sf2 and velocity V2 at the same station when the discharge Q2 is 200 cfs if the cross-sectional geometry is unknown. The at-a-station hydraulic geometry relationships (Table 5.3) can be used when no specific field data is available. For width:

æ Q2 ö W2 = W1 ç ÷ çQ ÷ è 1ø

0.26

æ 200 ö = 250 ç ÷ è 8000 ø

0.26

= 96 ft

For depth:

y o2

æ Q2 ö = y o1 ç ÷ çQ ÷ è 1ø

0.40

æ 1 ö =8ç ÷ è 40 ø

0.40

= 1.8 ft

Slope is unchanged:

S f1 = S f2 = 2.5 x 10 −4

For velocity:

æ Q2 ö V2 = V1 ç ÷ çQ ÷ è 1ø

0.34

æ 1 ö =5ç ÷ è 40 ø

0.34

= 1.4 ft / s

(b) Using the same station as for part (a), estimate the width, W 2. depth yo2, slope Sf2 and velocity V2 in an upstream section of this stream if the bankfull discharge is 500 cfs and the bed material is gravel (D50 = 8 mm). How would the hydraulic geometry change if the bed material upstream is sand (D50 = 0.6 mm)? The "downstream" geometry relationships can be used in this case. Two types of relationships are given in Table 5.3. The sand bed relationships are a function of discharge only, whereas the gravel bed relationships are a function of both discharge and sediment size. Both methods are compared in the following.

5.73

**Flow Depth Sand bed:
**

y o2

æ Q b2 ö = y o1 ç ÷ çQ ÷ è b1 ø

0.46

æ 500 ö =8ç ÷ è 8000 ø æ 1ö =8ç ÷ è 16 ø

0.46

= 2.2 ft

Gravel bed: Channel width Sand bed:

æ Q b2 ö y o2 = y o1 ç ÷ çQ ÷ è b1 ø

0.40

0.4

= 2.6 ft

æ Qb2 ö W2 = W1 ç ÷ çQ ÷ è b1 ø

0.46

æ 1ö = 250 ç ÷ è 16 ø

æ D s2 ç ç Ds è 1 ö ÷ ÷ ø

−0.33

0.46

= 70 ft

Gravel bed:

æ Q b2 ö W2 = W1 ç ÷ çQ ÷ è b1 ø

0.53

æ 1ö = 250 ç ÷ è 16 ø

0.53

æ 8 ö ç ÷ è 0.6 ø

−0.33

= 25 ft

**Friction slope Sand bed:
**

S f2

æ Q b2 ö = S f1 ç ÷ çQ ÷ è b1 ø

−0.46

= 2.5 x 10

−0.4

−4

æ 1ö ç ÷ è 16 ø

−0.46

= 9 x 10 − 4

Gravel bed:

S f2

æ Q b2 ö = S f1 ç ÷ çQ ÷ è b1 ø

æ D s2 ç ç Ds è 1

− 0. 4 ö 1ö æ 8 ö ÷ = 2.5 x 10 − 4 æ ç ÷ ç ÷ = 0.01 ÷ è 0 .6 ø è 16 ø ø

1

**Velocity Sand bed:
**

æ Q b2 ö V2 = V1 ç ÷ çQ ÷ è b1 ø

0.08

æ 1ö =5ç ÷ è 16 ø

æ D s2 ç ç Ds è 1 ö ÷ ÷ ø

0.08

= 4 ft / s

Gravel bed:

æ Q b2 ö V2 = V1 ç ÷ çQ ÷ è b1 ø

0.07

.33

æ 1ö =5ç ÷ è 16 ø

0.07

æ 8 ö ç ÷ è 0.6 ø

.33

= 9.7 ft / s

The gravel bed relationships give a steeper, faster flowing and narrower channel as compared to the sand bed relationships. Unless the sediment size is markedly different for two streams, the resulting hydraulic geometry calculated from both sets of equations will be similar. 5.9.7 PROBLEM 7 Downstream Sediment Size Distribution Measurements of sediment size in the St.-Lawrence Seaway between Cornwall and Valleyfield are given in the following table.

5.74

Table 5.9. Sediment Size Distribution in the St. Lawrence Seaway. D50 Distance Downstream of Cornwall (mm) (km) (mi) 0 0 28 24.1 15 0.25 40.2 25 0.018 48.3 30 0.003 56.3 35 0.001 Estimate the mean sediment size D50 10 miles (16.1 km) and 20 miles (32.2 km) downstream of Cornwall. The gradual decrease in sediment size with downstream distance can be approximated by the following equation: (SI) (English)

D 50 = 28 x 10 −0.082 x D 50 = 28 x 10 −0.132 x

This equation was obtained by regression analysis based on Equation 5.13. At distances of 10 and 20 miles (16.1 and 32.2 km), the expected mean sediment sizes D50 obtained by this relationship are, respectively, 1.34 mm and 0.064 mm. 5.9.8 PROBLEM 8 Scale Ratios for Physical Models (SI) A physical model is to be built in the Hydraulics Laboratory to simulate the flow pattern around 2 a structure in a complex multiple channel stream. About 334 m of space (with a maximum length of 18.3 m) is available in the laboratory to model a 800 m reach. Knowing that the same fluid (water) will be used for both the model and the prototype, determine the appropriate scale ratios for time, discharge and force. Also, required is the flow depth in the model at the location where flow depth reaches 6 m in the prototype. A fixed boundary model will be used and open channel flow modeling is scaled by similarity in Froude number. The scale ratios for γ and ρ equal unity and thus, scaling depends uniquely on -2 the length scale L = 18.3/(800) = 2.3 x 10 . The following scale ratios for time, discharge and force are calculated from the expressions shown in Table 5.7. Parameter Time Discharge Force Scale Ratio (Lρ/γ) L

5/2 1/2

=L

1/2

1/2

= 0.15

5/2

(γ/ρ)

=L

= 8.0 x 10

-5

-5

Lγ = L

3

3

= 1.2 x 10

The flow depth of the model at h = 6.0 m is given by the product hL = 0.14 m.

5.75

5.9.9 PROBLEM 9 Scale Ratios for Physical Models (English) A physical model is to be built in the Hydraulics Laboratory to simulate the flow pattern around 2 a structure in a complex multiple channel stream. About 3,600 ft of space (with a maximum length of 60 ft) is available in the laboratory to model a 1/2-mile reach. Knowing that the same fluid (water) will be used for both the model and the prototype, determine the appropriate scale ratios for time, discharge and force. Also, required is the flow depth in the model at the location where flow depth reaches 20 ft in the prototype. A fixed boundary model will be used and open channel flow modeling is scaled by similarity in Froude number. The scale ratios for γ and ρ equal unity and thus, scaling depends uniquely on -2 the length scale L = 60/(0.5 x 5280) = 2.3 x 10 . The following scale ratios for time, discharge and force are calculated from the expressions shown in Table 5.7. Parameter Time Discharge Force Scale Ratio (Lρ/γ) L

5/2 1/2

=L

1/2

1/2

= 0.15

5/2

(γ/ρ)

=L

= 8.0 x 10

-5

-5

L γ = L = 1.2 x 10

3

3

The flow depth of the model at h = 20 ft is given by the product hL = 0.45 ft.

5.76

these three publications provide a comprehensive integrated approach to analyzing stream instability and bridge scour problems and selecting and designing countermeasures for specific problems. 2). The Federal Highway Administration has prepared Hydraulic Engineering Circular (HEC) No.1 . The interested reader is referred to these two volumes for an analysis and assessment (Vol. 1994a. A treatise of great interest in relation to highway crossings is the Report FHWA-RD-78-162 and 163 on countermeasures for hydraulic problems at bridges by Brice and Blodgett (1978). selection. automobile bodies. While the methods for river training and bank stabilization discussed herein are applicable to short and long reaches of the river. Generally. It must also be recognized that the solution to a particular problem may generate problems elsewhere in the river system. restoring. training. and stabilizing rivers. and timber dikes. 6. angled and sloped rock-filled. and design guidelines for a wide range of stream instability and bridge scour countermeasures. brick. steel jack and brush jetties. sacked concrete and sand.1 OVERVIEW From a study of river morphology and river response (Chapter 5). they are not a panacea to all problems associated with highway encroachments on rivers. and bank stabilization of rivers associated with highway projects are confined to short reaches of the river. sheet and timber piles. and asphalt mattresses. It is not intended that an exhaustive coverage of the various types of river control structures and methods of design be made in this manual. willow. earth-filled. Concrete. restoration. and bank protection works. When used in combination with HEC-20 (Lagasse et al. Army Corps of Engineers (1981. The integrated and interactive effects of these structures with the river are discussed in Chapter 9.b) and the U.CHAPTER 6 RIVER STABILIZATION AND BANK PROTECTION 6. An early treatise on the subject of bank and shore protection was prepared by the California Division of Highways (1959). riprap grouted slope protection. 1) and 283 case histories (Vol. 2001) to provide experience. river cross section. rock. Numerous types of river control and bank stabilization devices have evolved through past experience. 23 (Lagasse et al. it should be clear that both short-term and long-term changes can be expected on river systems as a result of natural and man-made influences. rather. An understanding of river system dynamics is essential to selection. changes to river alignment. 2001) and HEC-18 (Richardson and Davis 2001). A large number of publications on river training and stabilization have been prepared by the U. Bureau of Reclamation. and concrete armor units have all been used in the practice of training. Recommended structures and design methods for river control are presented in this chapter.S. Many more publications on the subject exist in the open literature.S. design and successful installation of river stabilization. the purpose of this manual is to recommend methods and devices which provide useful alternatives to the highway engineer for the majority of circumstances which are likely to be encountered in highway practice.

there may also be regions within a river in which some degree of instability is exhibited for all flow conditions. Regardless of the fact that the majority of bank changes occur during comparatively short time periods. Changes in channel geometry with time are particularly significant during periods when alluvial channels are subjected to high flows. Bank retreat is often a combination of these processes and mechanisms varying at seasonal and sub-seasonal time scales. Other Subsurface Flows Seepage Forces Piping Waves Wind Boats 6. Table 1 lists factors affecting bank erosion. Climatic Factors Freezing Ice Thickness Duration Frequency Thawing Permafrost Precipitation F. bank height. A. Table 6. Recreational Boating Commercial Boating D. Raw banks may develop on the outside of bends as a consequence of direct impingement of the flowing water. Hydraulic Factors Fluid Properties Specific Weight Viscosity (Temp) Flow Characteristics Discharge Magnitude Duration Frequency Velocity Velocity Distrib.1. direct fluvial entrainment. 2001) provides more detail on these processes.2 . Turbulence Shear Stress Drag/Lift Forces Momentum Energy B. and mass failure.2. Human Factors Agricultural Activities Mining Dams Navigation Transportation Urbanization Drainage Floodplain Develop. In most instances when considering the instability of alluvial rivers. especially near-bank hydraulic fields. Factors Affecting Erosion of River Banks. and the geotechnical properties of the bank material. and/or retreat of a stream bank is dependent on the processes responsible for the erosion of material from the bank and the mechanisms of failure resulting from the instability created by those processes. The impact of these processes on bank retreat is dependent on site characteristics. 6.6. HEC-20 (Lagasse et al. Sloughing banks may occur as a result of seepage and other secondary forces created by water draining back through the banks into the river. Geomorphic Factors River Planform Meandering Straight Braided Anabranched Bed and Bank Material Size Gradation Shape Specific Weight Fall Velocity Bank Characteristics Cohesive Noncohesive Stratified Height C. it can be shown that approximately 90 percent of all river changes occur during the small percentage of the time when the discharge exceeds the dominant discharge.1 Causes of Streambank Failure Bank retreat processes may be grouped into three categories: weakening and weathering processes. generated either naturally or by human activities. Continuous wave action. instability. may also precipitate erosion problems. Biological Factors Vegetation Trees Shrubs Grass Animals Domestic Wild E. Erosive forces during high flow periods may have a capacity approximately 100 times greater than those forces acting during periods of intermediate and low flow. The converse situation exists during relatively dry periods.2 STREAM BANK EROSION The erosion.

Noncohesive bank material tends to be removed grain by grain from the bank line. because the duration of the change in stage is small. The reverse is true of the rivers that continuously gain water by an inflow through their banks. The rate of particle removal. The high water table may result from: (1) a wet period during which water draining from tributary watersheds saturates the floodplain to a higher level. frost heaving. the turbulent fluctuations. but may be partly protected by adjacent layers of cohesive material. With a rise in river stage an outward gradient is developed that induces flow into the banks.3 . Failure of banks for various situations is shown in Figure 6. The layers of noncohesive material are subject to surface erosion. However. such as may occur with hydropower operations. and cohesion. However. piping. is affected by factors such as the direction and magnitude of the velocity adjacent to the bank. (4) irrigated floodplains. seepage force. (3) increased infiltration resulting from changes in land use causing an increase in water level. the magnitude and fluctuations in the shear stress exerted on the banks. The inflowing water creates a seepage force that makes the banks less stable. a stabilizing seepage force is generated. permeability. such banks when undercut and/or saturated are more likely to fail due to mass wasting processes such as sliding. noncohesive. Rivers that continuously seep water into the banks tend to have smaller widths and larger depths for a particular discharge. flow will be from the banks into the river.2 Bed and Bank Material Resistance of a river bank to erosion is closely related to several characteristics of the bank material. piping and wave forces. the inflow and outflow phenomena are usually concentrated locally in the surface of the banks. (2) boat and wind waves which cause local variations in stage that introduce inflow and outflow of water from the banks. these types of banks consist of layers of materials of various sizes. and (5) development of the adjacent floodplain for homes and businesses that utilize septic tanks and leach fields to dispose of waste water and sewage. and subsurface flow on the stability of the banks. Cohesive bank material is more resistant to surface erosion and has low permeability which reduces the effects of seepage. (3) predominately dry and semi-arid channels subject to intermittent floods. (2) poor drainage conditions resulting from deterioration or failure of surface drainage systems. many of which may act concurrently. If the water table is higher than river stage. Composite or stratified banks are very common on alluvial rivers and generally are the product of past transport and deposition of sediment by the river. may exacerbate the bank erosion process. 6. Frequent stage fluctuations. and composite. More specifically.6. This can be caused by: (1) the storage and release of water for pumped storage hydropower generation which causes numerous fluctuations in river stage. and (3) the formation and loss of backwater caused by ice flows and ice jams which lead to both seepage into and out of the banks. and hence the rate of bank erosion. Bank material deposited in the river can be broadly classified as cohesive. This type of bank is also vulnerable to erosion and sliding as a consequence of subsurface flows and piping.3 Subsurface Flow With flow of water from the river into the adjacent banks.2. 6.2.1. The movement of water through the bank material can be attributed to various factors.

Typical bank failure surfaces (a) noncohesive.c).Figure 6. 6.b.1.4 . (b) cohesive. (c) composite (after Brown 1985a.

and increased infiltration of water into the floodplain as a result of changing land-use practices. The slope materials may then tend to move downward into the void in order to establish a new balance of forces or equilibrium. fluctuations in groundwater levels. For slope toes acted upon by erosive stream water. the material is slowly removed. Landslides. flow is induced in more permeable layers by changes in river stage and by wind-and boat-generated waves. operation of equipment on the floodplain adjacent to the banks. the slope loses more resistance by buttressing than it does by downslope gravitational forces.4 Piping of River Banks Piping is another phenomenon common to the alluvial banks of rivers. With stratified banks. The toe of the failed mass can provide a new buttress against further movements. an additional set of forces is added. result from an imbalance of forces.5 . the continual removal of toe material can upset the force balance. saturation of banks by leach fields from septic tanks. These forces are associated with the downslope gravity component of the slope mass. the downslope movement of earth and organic materials. added gravitational force resulting from tree growth. Bank erosion may continue on a grain-by-grain basis or the block of bank material may ultimately slide downward and outward into the channel. and vibration of the slope. piping. undermining portions of the bank. i. However. the force equilibrium may again be upset. A slope may fail if stable material is removed from the toe. When a slope is acted upon by a stream or river.The presence of water in the banks of rivers and its movement toward or away from the river affect bank stability and bank erosion in various ways. lenses of sand and coarser material sandwiched between a layer of finer cohesive materials.1c. 6. These forces are associated with removal of material from the toe of the slope. Various forces are involved in mass wasting. Resisting these downslope forces are the shear strength of the earth's materials and any additional contributions from vegetation via root strength or engineered slope reinforcement activities. location of roads that cause unfavorable drainage conditions. further reducing the stability of the affected block of bank material. and mass wasting. a block of bank material drops down and results in the development of tension cracks sketched in Figure 6. When the toe of a slope is removed.2.2. These cracks allow surface flows to enter. 6. 6. blocks of the bank may slump or slide into the channel. and mass wasting. Without this foundation material to support the overlying layers. If the flow through the permeable lenses is capable of dislodging and transporting particles from the permeable lenses. causing bank failure as a result of a combination of seepage forces. The related erosion of banks is a consequence of seepage forces.5 Mass Wasting An alternative form of bank erosion is caused by local mass wasting. Mass wasting may be further aggravated by construction of homes on river banks..e. Often this equilibrium is a slope configuration with less than original surface gradient. piping. if this buttress is removed by stream erosion. If the bank becomes saturated and possibly undercut by flowing water.

the river is shaped to an alignment consisting of a series of easy bends. and (9) protect longitudinal encroachments. thereby allowing a reduction in bridge length and height. 6. requiring additional work and possible abandonment of the original work. To minimize attack by the stream on stabilization and rectification structures.3. and control work on both banks through bridge crossings.2 Radius of Curvature The most appropriate radius of curvature for rectification and stabilization varies from river to river and from reach to reach for a given river.6 . fixed point on the bank and continues downstream to another stable location or to some point below which the river can safely be left uncontrolled. or to repair damage and improve initial designs. or will shift the attack to some other nearby reach of bank. thereby reducing backwater and scour and facilitating passage of ice and debris.8. It must be determined on the basis of relatively stable natural bends for each stream (see Section 5. (6) permit construction of a well-aligned bridge crossing by diverting the channel from a skewed alignment. Revetments should be constructed on a smooth alignment. minor work on convex bars.4). (8) secure existing works. in order to avoid eddies set up by disturbances to the flow that can lead to local scour and subsequent undermining of the revetment. The optimum bend radius approximates that of relatively stable bends in the general river reach. with no irregularities.8. insofar as practicable.1 Fixed Points One of the essential requirements in designing a system of stabilization works is that construction starts at a stable.3).6. (7) reduce the overall cost of a road project by diverting the channel away from the base of a valley slope. Construction of relatively short isolated stabilization work has often proved unsuccessful because eventual changes in the direction of flow inherent in bank caving in the upstream uncontrolled reach either will set up a direct attack against the isolated protective work and severely damage or destroy it. (2) economize on bridge lengths by constricting the natural waterway. Specific functions of bank protection and training works in relation to bridges and their approaches include: (1) stabilize eroding river banks and channel location in the case of shifting streams. (3) direct flow parallel to piers and thereby minimize local scour. because there is a tendency for flows to shift from side to side in such reaches. A comprehensive bank stabilization and channel rectification program to control a river reach completely normally requires extensive work on concave banks in bends. (4) improve the hydraulic efficiency of a waterway opening. aided in some instances by hydraulic model studies. 6.3.3 RIVER TRAINING AND STABILIZATION Various devices and structures have been developed to control river flow along a preselected path and to stabilize the banks. 6. with the flow directed from one bend into the next bend downstream in such a way as to maintain a direction essentially parallel to the channel flow line (see Section 5. Straight reaches and reaches of very small curvature should be avoided. (5) protect road approaches from stream attack and prevent meanders from migrating into the approaches. Most have been developed through trial and error applications.

Rigid revetments are typically impermeable and do not have the ability to conform to changes in the supporting surface. but rather. 6. To protect against lateral channel instability. River training structures are distinctive in that they alter hydraulics to mitigate undesirable erosional and/or depositional conditions at a particular location or in a river reach.7 . River training structures are described as transverse. River training structures can be constructed of various material types and are not distinguished by their construction material. River training structures are those which modify the flow (flow control). longitudinal or areal depending on their orientation to the stream flow. In the final report to Congress. The deeper the channel is. Revetments can be classified as either rigid or flexible/articulating.S.The shorter the radius of curvature of a bend. These countermeasures often fail by removal and displacement of the armor material.3. They are usually applied in a blanket type fashion for areal coverage. flow control structures are used to: • • • • Direct flow from one bend into the next bend downstream Flair out sharp bends to a larger radius of curvature to provide a more desirable channel alignment Close off secondary channels and old bendways Concentrate flow on a limited width within a wider channel To protect against vertical channel instability. the greater the possibility of undermining bank protection work in the bend and the greater the cost of maintaining the structure. and other advantages.3 Countermeasures for Channel Instability Countermeasures can be used to control both lateral and vertical channel instability and include river training structures and revetment armoring. Rock riprap is probably the most widely used revetment material to stabilize river banks and protect the side slopes of embankments and river training devices. because of durability. Army Corps of Engineers (1981) concluded that rock will likely continue to be the first choice of bank protection materials where material of sufficient size is available and affordable. These countermeasures often fail due to undermining. Flexible/ articulating revetments can conform to changes in the supporting surface and adjust to settlement. the U. Therefore. but their orientation to flow. flow control structures are used for: • • • Limiting or halting long-term degradation Establishing a desired channel bed elevation in a bridge reach Arresting the migration of a head cut or nickpoint through a bridge reach Revetments are structures parallel to the current and are used to armor the channel bank from erosive/hydraulic forces. sharp curvature of bends should be avoided to obtain the most economical control of the river. Because of its wide use and 6. the deeper the channel will be adjacent to the concave bank.

6. filters. the general level of maintenance resources required. grouted rock. erosion mechanisms and riprap failure modes. including river training devices and revetment armoring. and design guidance for a wide range of stream instability and bridge scour countermeasures. and biotechnical engineering approaches. hydraulic analysis. preformed blocks.8 . Design concepts included in HEC-11 are: design discharge. • • • Specific design guidelines are provided in HEC-23 for the following river training devices: • • • • Bendway Weirs/Stream Banks Spurs Guide Banks Check Dams/Drop Structures Design guidelines are also provided for the following armoring (revetment) countermeasures. environmental permitting. special design considerations related to riprap. Provide detailed design guidelines for specific countermeasures. HEC-23 is organized to: • • Highlight the various characteristics. and which State Highway Agencies (SHAs) have experience with specific countermeasures. HEC-11 includes discussions on erosion potential. • • • • • Revetment (summary of HEC-11 guidance) Soil Cement Wire Enclosed Riprap Mattress Articulating Concrete Block Systems Articulating Grout Filled Mattress Hydraulic Engineering Circular (HEC) No.4 Countermeasure Design Guidelines Hydraulic Engineering Circular (HEC) No. flow types. list information on their functional applicability to a particular problem. 2001) provides experience. 23 (Lagasse et al.6) is devoted to design and placement of rock riprap. Discuss countermeasure design concepts including design approach.3. extent of protection. a separate section (Section 6. as well as riprap types including rock riprap. flow resistance. and toe depth. and paved linings. gabions. selection. their suitability to specific river environments. groups of countermeasures and identify their individual For specific countermeasures. channel geometry. 6. Provide general criteria for selection of countermeasures for bridge scour and stream instability problems. rubble riprap. 11 (Brown and Clyde 1989) is a comprehensive design manual for riprap revetment. and edge treatment. Detailed design guidelines are presented for rock riprap.importance in highway practice.

and design procedures are summarized in charts and examples.7. excessive piping through the filter can result in erosion of the subgrade beneath the armor. If openings in the filter material are too large. should be designed with an apron or toe trench so that the system provides protection below the combined expected contraction scour and long-term degradation depth. Variations in bed elevation during flow events or after bank hardening can result in the undermining of bank protection structures including longitudinal structures. prevent migration of fine soil particles through voids in the armoring material. Conversely. hydrostatic pressures can build up in the underlying soil and result in failure of the countermeasure. 6. if openings in the filter are too small. 2001). HEC-23 (Lagasse et al. design. and specifications of filter material can be found in HEC-11 (Brown and Clyde 1989). a launching apron can be used to provide enough volume of rock to launch into the channel while maintaining sufficient protection of the exposed portion of the bank. Deep sections at the toe of the outer bank of a bendway are the result of scour. especially armoring countermeasures such as bankline revetment or armoring used as protection for river training works such as spurs and guide banks.5 deals with riprap design and placement. high velocity along the outer bank is caused by secondary currents and greater outerbank depths. When excavation to the contraction scour and degradation depth is impractical.9 . and together with the resultant shear stress. produce scour and cause a 6. For revetment slope protection.8. Section 6. Experience has indicated that the proper design of filters is critical to the stability of revetments.7 and issues related to overtopping flows on embankments are summarized in Section 6.4. toe. Environmental considerations for streambank protection are presented in Section 6.3. As discussed in Section 5. and Section 6. Undermining of the edges of armoring countermeasures is another of the primary mechanisms of failure. Design guidance is also presented for: • • • Wire-enclosed rock (gabions) Precast concrete blocks Concrete paved linings The following section (6. Filters prevent soil erosion beneath the armoring material.9. Continuous systems. and permit relief of hydrostatic pressure within the soils.4) summarizes design concepts for selected flow control (river training) structures. Guidelines for the selection.5 Protection of Training Works Granular or geosynthetic filters are essential to the performance of hydraulic countermeasures. Additional guidelines on edge treatment for armoring countermeasures can be found in HEC-11 and HEC-23. and Section 6. General filter design concepts are discussed in Section 6. The edges of the armoring material (head. Tension anchors may be used to increase stability at the edges of these continuous systems. and flanks) should be designed so that undermining will not occur.6 describes design for revetment types other than riprap. distribute the weight of the armor units to provide a more uniform settlement. such as articulating concrete block systems and grout filled mattresses applied on side slopes.3. this is achieved by trenching the toe of the revetment below the channel bed to a depth which extends below the combined expected contraction scour and long-term degradation depth.

hardening of the outer bank by longitudinal bank protection structures may cause the channel cross section to narrow and deepen by preventing the recruitment of eroded outer bank sediments. Although scour-depth can be estimated analytically or empirically. empirical methods were generally found to provide better agreement with observed data. Lacking experience on a particular stream. velocity. Since secondary currents transport sediment supplied. 6.4 FLOW CONTROL STRUCTURES A flow control structure is defined here as a structure. Experience is usually the most reliable means of estimating scour depth when designing a bank protection project for a particular stream. a spur or a series of spurs may protect the stream bank more effectively and at less cost than revetment riprap applied directly on the bank. Structures made of riprap. and cannot be applied to channels that have been confined to widths significantly less than a natural system. 6. Also. the techniques presented by Maynord are restricted to meandering channels having naturally developed widths and depths. failure of riprap on the bank may immediately endanger adjacent structures. HEC-23 also contains guidance for estimating scour at vertical wall structures (e. and to establish a more desirable channel alignment or width. or the meshes of wire. or depth of flowing water. Maynord (1996) provides an empirical method for determining scour depths on a typical bendway bank protection project. failure of the riprap on the spur can often be repaired before damage is done to structures along and across the river. but these are classed as impermeable because they act essentially as impermeable barriers to a rapidly moving current of water. or filled with riprap. such as openings between adjacent boards or pilings. By deflecting the current from the bank and causing sediment deposits behind them. Structures within this category are sometimes called "river training works". HEC-23 provides application guidelines for Maynord's approach to estimating toe protection requirements on stabilized bendways. Although his studies are restricted to sand bed streams. have some degree of permeability.2. from outer bank erosion toward the inner bank of a bend.4. As used here. Among the most important properties of a flow control structure is its degree of permeability. retards and bulkheads) from flow parallel to and impinging on the wall. in large part. by moving the location of any scour away from the bank.g. scour depths may be estimated using physically based analytical models or empirical methods.difference between the sediment load entering and exiting the outer-bank zone. whereas a permeable structure may serve mainly to reduce water velocity. 6. An impermeable structure may deflect a current entirely. Nonetheless. Types of flow control structures are distinguished on Figure 6. either within or outside a channel that acts as a countermeasure by controlling the direction. the Maynord method agrees reasonably well with the limited number of gravel-bed data points obtained by Thorne and Abt (1993)..1 Spurs A spur is a structure or embankment projected into a stream from the bank at some angle and for a short distance to deflect flowing water away from critical zones. the term "permeable" means that a structure has definite openings through which water is intended to pass. to prevent erosion of the bank.10 . Conversely.

This decreases the cost of the bridge structure. The use of spurs in this instance may decrease the length necessary for the bridge opening and may make a more suitable. thus maintaining its location from year to year.11 . 2001). Spurs on streams with suspended sediment discharge can cause deposition to establish and maintain the new alignment. poorly defined stream into a well-defined channel that neither aggrades nor degrades. Recommendations for spur design from Brown (1985) are summarized in HEC-23 (Lagasse et al. Spurs constructed perpendicular to the highway embankment keep the potentially erosive current away from the embankment. Spurs. and groins. and jack fields may be either upstream or downstream from the bridge (from Brice and Blodgett 1978). thus protecting it.Figure 6. dikes. Spurs are also used to protect highway embankments that form the approaches to a bridge crossing. which are also used to describe these structures. crossing. retards. Placement of flow control structures relative to channel banks. Spurs are also used to channelize a wide. The major considerations are: • • • • • • • • Extent of Channelbank Protection Spur Length Spur Spacing Spur Angle/Orientation Spur Height Spur Crest Profile Channel Bed and Channel Bank Contact Spur Head Form 6. jetties. stable channel approach to the bridge. and floodplain.2. Spurs as used in this report encompass the terms dikes. Often these highway embankments cut off the overbank flood flows causing these flows to run parallel to the embankment enroute to the bridge opening.

bank barbs.3. 6. Design guidelines for bendway weirs are presented in HEC-23 (Lagasse et al.6. steel or timber. straight reaches not subject to direct attack. and reverse sills.3). The structures protrude only short distances into the river channel and are supplemented with a root section extending landward into the bank to preclude flanking. The design of timber pile retards is essentially the same as timber pile dikes shown in Figure 6. and the upper part is covered with topsoil and seeded with native vegetation. Figure 6.2 and 6.c). They may be used in combination with bank protection works such as riprap. 2001).2 Bendway Weirs Bendway weirs. Perspective of hard point with section detail (after Brown 1985a. 6. The structures are especially adaptable in long. Bendway weirs are used for improving inadequate navigation channel width at bends on large navigable rivers. or riprap can be eliminated. The retard then serves to reduce the velocities sufficiently so that either smaller riprap can be used. also referred to as stream barbs. are low elevation stone sills used to improve lateral stream stability and flow alignment problems at river bends and highway crossings.4. Pile retards can be made of concrete.4. They are used more often for bankline protection on streams and smaller rivers. should excessive erosion persist.6.4.4 Retards Retards are devices placed parallel to embankments and river banks to decrease the stream velocities and prevent erosion (Figures 6.3 Hardpoints Hardpoints are an erosion control technique consisting of stone fills spaced along an eroding bank line (Figure 6.12 .4). 6.b. The majority of the structure cannot be seen as the lower part consists of rock placed underwater.

Figure 6. Timber and concrete cribs are sometimes used for bulkheads and retaining walls to hold highway embankments.5. 6. Concrete or timber cribs.Figure 6. drainage holes (weep holes) must be provided. However. The crib may be slanted or vertical depending on height and the crib is filled with rock or earth.5. The foundation of these walls should be placed below expected scour depths. Cribs are made up by interlocking pieces together in the manner shown in Figure 6. Retard. In constructing concrete retaining walls. Reinforced concrete retaining walls are alternatives to timber cribs which can be considered.4. particularly where lateral encroachment into the river must be limited. concrete retaining walls are expensive and are generally only used in special confined locations where space precludes other methods of bank protection.13 .

These are similar in function to guide banks. The arrangement of piles depends upon the velocity of flow.5 Dikes (Floodplain) A floodplain dike is an impermeable linear structure for the control or containment of overbank flow. 1981. encourage diversity of various channel depths. the tops of which are constructed below the design water surface elevation and would not connect to the high bank. and gap length U. Timber or steel pile dikes (also retards) may consist of closely-spaced single. and rocks may be used to fill the space between the piles. flow depth and available pile lengths.4. Permeable dikes are those which permit flow through the dike but at reduced velocities. The various forms of pile dikes are illustrated in Figure 6. Pile dikes are vulnerable to failure through scour. although timber pile dikes have been used in the Columbia River. The length of each dike depends on channel width. Generally. but are usually much longer and may extend to the valley side. The findings from a model investigation of these structures include the effects of various vane dike orientation. 1978). The structures will discourage high erosive velocities next to an unprotected bank line. or to confine channel width and maintain channel alinement. quantity of suspended sediment transport. 6. the permeable pile dikes will only partially be effective in training the river and protecting the bends. Army Corps of Engineers. Deposition of suspended sediments in the pile dike field is a necessary consequence of reduced velocities. Such dikes are commonly called "training dikes. The spacing between dikes varies from 3 to 20 times the length of the upstream dike. pile dikes are not used in large rivers where depths are great. Some dikes extend upstream from one or both sides of the bridge opening. or multiple rows.6. This can be overcome if the piles can be driven to a large depth to achieve safety from scour. For example. On the other hand. If there is not sufficient concentration of suspended sediment in the flow. these dikes can be quite effective.4. Floodplain dikes are used to prevent flood water from bypassing a bridge.6 Dikes (Channel) Both permeable and impermeable dikes are installed in channels. Most dikes are on floodplains (Figure 6. Vane dikes are low-elevation structures designed to guide the flow away from an eroding bank line (Figure 6.14 . There are a number of variations to this scheme. Water would be free to pass over or around the structure with the main thread of flow directed away from the eroding bank. in moderate flow velocities with high concentrations of suspended sediments. banks of wide shallow rivers can be successfully protected with dikes. with closer spacing favored for best results.4.S.6. position relative to other dikes. double. Stabilization of the bank by other methods should be considered. or the velocities in the dike fields are too large for deposition. vane dike length. 6. The structures can be constructed of rock or other erosion-resistant material. Double rows of piles can be placed together to form cribs. wire fence may be used in conjunction with pile dikes to collect debris and thereby cause effective reduction of velocity. and depth and width of the river." The use of fencetype structures for training dikes is mostly restricted to arid and semi-arid regions (Brice et al. On the other hand. thereby preventing further erosion of the banks and causing deposition of suspended sediment from the flow. If the velocity of flow is large.7). pile dikes are not likely to be very effective. and protect existing natural bottomland characteristics.2) but in some situations (as on wide braided rivers or on alluvial fans) they may be within channels (see Section 6.6). or the base of the piles can be protected from scour with dumped rock in sufficient quantities.

7. Army Corps of Engineers 1981). Vane dike model.Figure 6.15 . during low stage portion of test run (U. Pile dikes (retards would be similar). ground walnut shell bed.6.S. 6. Figure 6.

Both lateral and longitudinal rows of jacks are used to make up the jetty field as shown in Figure 6.4. A typical layout is shown in Figure 6. there must be considerable debris in the stream to collect on the fence and the suspended sediment concentration must be large so that there will be deposition behind the retard. Two types are shown in Figure 6.7 Jetties The purpose of a jetty field is to add roughness to a channel or overbank area to train the main stream along a selected path.6. 6. it may be necessary to use jetty fields on both sides of the river channel because in flood stage the river may otherwise develop a chute channel across the point bar.8. When jetty fields are used to stabilize meandering rivers.8. Figure 6. Steel jacks are devices with basic triangular frames tied together to form a stable unit. Tiebacks should be spaced every 30 m (l00 ft) and space between jacks should not be greater than their width. depending upon the debris and sediment content in the stream.8. The resulting framework is called a tetrahedron. and may be 15 to 75 m (50 to 250 ft) apart.16 . The lateral rows are usually angled about 45 to 70 degrees downstream from the bank. Various forms of steel jacks may be assembled. In order to function well. Typical jetty-field layout. Wire fencing may be placed along the row of tetrahedrons. The added roughness along the bank reduces the velocity and protects the bank from erosion. The tetrahedrons are placed parallel to the embankment and cabled together with the ends of the cables anchored to the bank.9. The spacing varies. Jetty fields are usually made up of steel jacks tied together with cables. Jetty fields are effective only if there is a significant amount of debris carried by the stream and the suspended sediment concentration is high.

Concrete. steel. however. Fill material placed between the bulkhead and natural bank should be free draining so that the soil behind the bulkhead will not become saturated and push the structure over. The bulkhead should be tied into the bank at the upstream and downstream end of the structure to prevent flow behind the bulkhead.8 Bulkheads Bulkheads can be used to prevent streambank erosion or failure. weepholes should be drilled in the fence at regular intervals to allow the bank to drain. aluminum. Timber bulkhead construction is similar to common fence construction except that a few precautions should be observed: • • All wood should be treated with preservative to minimize deterioration due to repetitive wetting and drying or insect activity. 6. As an additional benefit. a bulkhead may provide a substantial increase in waterfront area and an improvement in water/land access.17 . • • • • 6.4. The toe of the bulkhead should always be protected with riprap. the service life is longer and less maintenance is required. A filter must be properly designed to match the filter with the soil. If there are no cracks between the planks.9. Timber is the most commonly available material for economical bulkhead construction. followed by the structure tipping over due to the pressure of the bank behind the bulkhead. Concrete and steel bulkheads generally cost at least four time as much as a comparable bulkhead of another material.Figure 6. Piles should be anchored to deadmen buried in the bank. corrugated asbestos. Steel jacks. The most common cause of bulkhead failure is scour around the pilings. Filter fabric or gravel can be placed as a filter behind openings in the fence to prevent fine soils from leaching through. and used tires have been used to construct bulkheads. timber and more recently.

4. and size of bridge opening.6. They are also used to protect the highway embankment and reduce or eliminate local scour at the embankment and adjacent piers. 6.10 Guide Banks Guide banks are placed at or near the ends of approach embankments to guide the stream through the bridge opening. such as eddies and cross-flow. A model investigation and literature review of longitudinal fence retards with tiebacks were conducted to identify the following important design considerations: • • Channel gradient must be stable and not be steep (tranquil flow) Toe scour protection can be provided by extending the support posts well below the maximum scour expected or by placing loose rock at the base of the fence to launch downward if scour occurs at the toe Tiebacks to the bank are important to prevent flanking of the fence and to promote deposition behind the fence Fence retards generally reduce attack on the bank so that vegetation can establish Metal or concrete fences are preferred due to ice damage and fire loss of wooden fences • • • 6. Both longitudinal (parallel to stream) fence retards and transverse (perpendicular to stream) fences have been used in the prototype with varying degrees of success.9 Fencing Fencing can be used as a low-cost bank protection technique on small to medium size streams. A typical guidebank at the end of an embankment is shown in Figure 6. Typical guidebank.10. flow disturbances. The effectiveness of guidebanks is a function of river geometry. Figure 6. quantity of flow on the floodplain.10. Special structural design considerations are required in areas subject to ice and floating debris. Constructed properly.4.18 . will be minimized to make a more efficient waterway under the bridge.

The crest elevation of guidebanks should be higher than the elevation of the design flood taking into consideration the effect of the contraction of the flow. or treated timber piles.12. the height of drop required. Shorter guide banks may be used where floodplain flow is small. The length of the guide bank. Ls. the guidebank should be placed at an angle with respect to the embankment with the view of streamlining the flow through the bridge opening.11 and 6. for skewed crossings. 2001). The major axis should be approximately parallel to the main flow direction. Design considerations for vertical wall or sloping sill structures are given in texts by Rouse (1950). sheet piles. 6.4. Peterson (1986). width of bridge opening and skewness of the highway crossing. Design guidelines for a vertical drop structure and stilling basins for drop structures are given in HEC-23 (Lagasse et al. 2001). required depends upon quantity of flow on the floodplain.The recommended shape of a guidebank is a quarter ellipse with a major to minor axis ratio of 2. The material used to construct the structure depends on the availability of materials.11 Drop Structures Check dams or channel drop structures are used downstream of highway crossings to arrest head cutting and maintain a stable streambed elevation in the vicinity of the bridge. Chow (1959). this is because the design flow should not overtop the guidebank. For bridge crossings normal to the river. Definition sketches for a vertical wall and a sloping sill drop structure are shown in Figures 6. or where trees or brush are intersected by the guidebank.19 . 2001). concrete. gabions. Design guidelines and a design chart for guide banks are provided in HEC-23 (Lagasse et al. However. the major axis would be normal to the highway embankment. Check dams are usually built of rock riprap. and the width of the channel. 6. and Simons and Senturk (1992).5. scour potential at piers and embankment ends is small.11. Definition sketch for a vertical drop (Lagasse et al. Figure 6.

12. Construction is not complicated so for many applications special equipment or construction practice is not necessary. Wave runup on rock slopes is usually less than on other types of bank protection.1 Factors to Consider When available in sufficient size. 6. Flow and scour patterns at a sloping sill (after Laursen and Flick 1983). The important factors to be considered in designing rock riprap bank protection are: • • • • • • • • • Durability of the rock Density of the rock Velocity (both magnitude and direction) of the flow in the vicinity of the rock Slope of the bed and bankline being protected Angle of repose for the rock Shape and angularity of the rock What shape and weight of stones will be stable in the streamflow What blanket thickness is required Is a filter needed between the bank and the blanket to allow seepage but to prevent erosion of bank soil through the blanket 6.20 . Riprap is usually durable and recoverable and may be stockpiled for future use. there are no special foundation requirements. Riprap stability increases with increasing thickness as more material is available to move to damaged areas and more energy is dissipated before it reaches the filter and streambank. Although riprap must be placed to the proper level below the bed.Figure 6. The cost-effectiveness of locally available riprap provides a viable alternative to many other types of bank protection. Rock riprap has many other advantages over other types of protection. rock riprap is usually the most economical material for bank protection. The appearance of rock riprap is natural and after a period of time vegetation will grow between the rocks.5 RIPRAP DESIGN AND PLACEMENT 6. Local damage or loss is easily repaired by the placement of more rock. A riprap blanket is flexible and is neither impaired nor weakened by slight movement of the bank resulting from settlement or other minor adjustments.5.

and (3) the characteristics of the rock including the geometry. the stability of rock riprap particles on a side slope is a function of: (1) the magnitude and direction of the stream velocity in the vicinity of the particles. 6. The drag force is normal to the lift force and is zero when the fluid velocity is zero. Diagram for riprap stability conditions. The fluid forces on a rock particle identified as P in Figure 6. The point of contact about which rotation in the R direction occurs is identified as point "0" in Figure 6.2 Stability Factor Design Methods Stability Factors For Riprap. This development closely follows that given by Stevens and Simons (1971). The drag force Fd is defined as the fluid force acting on the particle in the direction of the velocity field in the vicinity of the particle. so it is appropriate to consider the stability of rock particles in terms of moments about the point of rotation. In the absence of waves and seepage.13a result primarily from fluid pressure around the surface of the particles.13. angularity and density. The functional relations between the variables are developed below. Consider flow along an embankment as shown in Figure 6.• • How will the blanket be stabilized at the toe of the bank How will the blanket be tied into the bank at its upstream and downstream ends 6.13.5. The lift force is zero when the fluid velocity is zero.13b the direction of movement is defined by the vector R .13c. The lift force F4 is defined herein as the fluid force normal to the plane of the embankment. (2) the angle of the side slope. In Figure 6. The remaining force is the submerged weight of the rock particle W s.21 . Figure 6. Rock particles on side slopes tend to roll rather than slide.

4. = e 2 Ws cos θ e1Ws sin θ cos β + e 3Fd cos δ + e4Fl (6. = cos θ tan φ η' tan φ + sin θ cos β (6. This analysis shows that the stability factor for rock riprap on side slopes where the flow has a non-horizontal velocity vector is related to properties of the rock.13c.1) The moment arms e1. e3 and e4 are defined in Figure 6. there is a balance of moments about the point of rotation such that e2 Ws cos θ = e1Ws sin θ cos β + e3Fd cos δ + e4Fl (6.22 . the rock is at the condition of 6.5) and ì1 + sin (λ + β) ü η′ = η í ý 2 î þ (6.6) Given a rock size Ds.4) (6.F. At incipient motion. the set of four equations (Equations 6. The angle θ is the side slope angle.2) The following particle stability analysis was first derived by Stevens (1968).F. of specific weight Ss and angle of repose φ and given a velocity field at an angle λ to the horizontal producing a tractive force τo on the side slope of angle θ. Accordingly.13b.13c and the angles δ and β are defined in Figure 6.F. against rotation of the particle is defined as the ratio of the moments resisting particle rotation out of the bank to the submerged weight and fluid force moments tending to rotate the particle out of its resting position.F.13b. 6. side slope and flow by the following equations: S. If S. S.5. and 6.3.The forces acting in the plane of the side slope are Fd and W s sinθ as shown in Figure 6.F. is unity. 6. is greater than unity. e2. The lift force acts normal to the side slope and the component of submerged weight W s sinθ acts normal to the side slope as shown in Figure 6. the riprap is stable. The stability factor S.3) in which ü ì ï ï cos λ ï ï β = tan −1 í ý ï 2 sin θ + sin λ ï ï ï þ î η tan φ η= 21τ o (S s − 1) γD s (6.6) can be solved to obtain the stability factor S. The analysis is also presented in Simons and Senturk (1992).F. if S.

incipient motion.e. æ η tan φ ö β = tan −1 ç ç 2 sin θ ÷ ÷ è ø (6. if S. Simplified Design Aid For Side Slope Riprap. the expression for the stability factor for horizontal flow on a side slope is: S.4 and 6.10 for η. the riprap is unstable. When the velocity along a side slope has no downslope component (i.) S m cos θ (6.8) When Equations 6.) 2 Sm 2 (S.7) and æ 1 + sin β ö η′ = η ç ÷ 2 ø è (6. the equations relating the stability factor.6 with λ = 0. then: η= 2 − (S. and the angle of repose for the rock are obtained from Equations 6. Here.9) in which ζ = S m η sec θ (6. 6.3. = Sm 2 {ζ 2 +4 −ζ } (6. Problem F. the specific weight of the rock is taken as 2. = 1. For horizontal flow along a side slope.14.8 are substituted into Equation 6.10) and Sm = tan φ tan θ (6.11) If we solve Equations 6.23 .F. the side slope angle.F.12) The interrelation of the variables in these two equations is represented in Figure 6.5) is the result of studies of the riprap embankment model data obtained by Lewis.F.F. is less than unity. This recommended stability factor for the design of riprap (S. the velocity factor is along the horizontal)..5 is employed.9 and 6.1 in Appendix F illustrates how to determine the stability of riprap.7 and 6.65 and a stability factor of 1. some simple design aids can be developed. the stability number. These studies were reported by Simons and Lewis (1971).F.

6.3 Velocity Profile and Tractive Force In riprap design.5. (2) Select a side slope angle.14.14.5 stability factor for horizontal flow along a side slope. the reference velocity Vr at a distance D50 above the bed is determined from Equation 6. it is often desirable to relate the tractive force (shear stress) acting on the riprapped bed or bank to the fluid velocity in the vicinity of the riprap. (3) Compute Sm from Equation 6.5. the riprap size required is obtained from: Dm = 21τ o (S s − 1) γ η (6.14.5) (2.Figure 6.12 with S. 6. = 1. then. For fully turbulent flow. For example θ = 25°.31 ÷ ø æ (2.14) 2 − (1. tan 25° (4) Compute η from Equation 6.13) where the stability number η is obtained from Figure 6.5 ) 2 η=ç ç (1.24 . Stability numbers for a 1.14 are computed in the following manner: (1) Select an angle of repose φ.11: Sm = tan 45° = 2.5 ö ÷ cos 25° = 0. from Equation 6. For example φ = 45°.14. If the shear stress τo on the side slope is known.14 ) 2 è (5) Repeat the above steps for the full range of interest for φ and θ. F. The curves in Figure 6.

and coefficients for a factor of safety (Sf ). One can also demonstrate (Richardson et al.5) η= 0.52 V* = 8.15) The relation is valid only for uniform flow in wide prismatic channels in which the flow is fully turbulent.25 . 1975) that the reference velocity Vr is related to the velocity against the stone Vs (Vr _ 1.52 ρ ø (6.15 can be used when the flow is accelerating such as on the tip of spur dikes or abutments. Equation 6. blanket thickness (CT ).14) Thus. stability (Cs).60 Vs2 (S s − 1) g D s (6.4 V ).5.5 V* ln ç ç 30.S.16) η= (6.30 Vr2 (S s − 1) g D s 0. In these areas.17a) 2 ì ï ï η = 0. For the purpose of riprap design. Army Corps of Engineers EM 1110-2-1601.30 í ï ln ï î 3 .æ D 50 Vr = 2. local depth of flow (y). the relation between Vr and τo is: τo = ρ Vr2 72 (6.S.4 U. Army Corps of Engineers Design Equation U. the shear stress is larger than calculated for Equation 6. s In summary the following expressions for η are equivalent: η= 21 τ o (S s − 1) γD s (6. The principal equation determines the riprap size of which 30 percent is finer by weight (D30) (U.17b) 6.3 D ) ÷ ÷ s ø è ü ï ï ý ï ï þ V2 (S s − 1) gD s (6. Army Corps of Engineers design equations are based on local depth-averaged velocity (V). vertical velocity distribution (Cv ). The equation is: 6. The equation should not be used in areas where the flow is decelerating or below energy dissipating structures.4 æ yo ö ç ç (12. and side slope (K1).S.2 D 50 è ö τo ÷ ÷ = 8. 1991 and 1994a and Maynord 1988).15.

5 or D50 (max). to highly nonuniform exhibited the same stability if they had the same D30. lb/ft Local depth-average velocity. Kg/m .81 m/s .283 – 0. ft Safety factor.36 for rounded rock 1/3 Gradation uniformity coefficient (D50 = D30 (D85/D15) Vertical velocity distribution coefficient 1. (D85/ D15 = 1.5 (6. m. Its use probably results in an overall increase in the sizes of the riprap in the blanket compared to using the traditional D50 approach.2) 0.” The Corps manual gives an approximate relationship between D50 and D30. ft/s Average velocity in the main channel. vessels. minimum 1. inside of bends 1.1 Stability coefficient for incipient failure. ft Blanket thickness coefficient 1. ft 3 3 Unit weight of water.D 30 ù éæ γ ö 1 / 2 V ÷ = S f C s C v C T y êç ç γ − γ ÷ (K gy )1 / 2 ú ú ê 1 ø û ëè s 2 . ft/s Local depth average velocity on side slopes. whichever is greater Local depth of flow. Natural variations in the quality of the rock used as riprap Vandalism Placement quality Freeze-thaw of the riprap is anticipated Precision of the determination of the hydraulic variables 6. ft Water-surface width at upstream end of bend (main channel only) m.25 downstream of concrete channels and at ends of dikes Centerline radius of curvature of bend (main channel only) m.7 to 5. Army Corps of Engineers manual (1994a) states that the minimum safety factor may have to be increased for the following conditions: • • • • • • Impact forces from logs. ice etc.S.2 log (R/W) for outside of bends. 32. ft/s Side slope correction factor 2 2 Acceleration of gravity. showed that gradations ranging from uniform. m/s. Kg/m .18) where: D30 Sf Cs = = = = = = = = = = = = = = = = = = = = = = Riprap size. m/s.2 ft/s D85/D15 Cv R W CT y γ γs V Vss Vavg K1 g The use of D30 for the sizing of the riprap is somewhat controversial.26 .30 for angular rock 0. 1 for (R/W) > 26 1. uprooted trees. Maynord (1988) states ”Stability tests conducted at a thickness of 1 X D100. m. which is the most commonly used thickness for bank protection. lb/ft 3 3 Unit weight of rock.0 for straight channels.0 for thickness = 1 D100 (max) or 1. The U. 9. m/s. It 1/3 is D50 = D30 (D85 / D15 ) .

52 Log ç ÷ Vavg èWø (6. Also.21) EM 1110-2-1601 states Equation 6. Determining the depth-averaged velocity to design riprap to protect side slopes is more difficult because the velocity varies greatly form the toe to the top of the bank. Vss æRö = 1. quality control. This velocity can be determined by physical or computer models. For this reason. The equation is: D 30 = 1. revetment top protection and revetment end protection (see also HEC-23). He describes a computer program “CHANLPRO” available from the U.21 is applicable for thickness = 1. the side slope correction factor in Equation 6.95 S 0. 3 unit weight of 167 lb/ft .19) Note that the value of Vss / Vavg rarely exceeds 1. Army Engineer Waterways Experiment Station that incorporates the above riprap design and scour depth procedures as well as sizing gabion mattresses.18.18. slopes from 2 to 20 percent and uniform flow with no tailwater. end protection is needed. is taken from the Carter et al. The Corps manual (1994) provide the following additional guidance. use the local depth-averaged velocity and local depth in Equation 6. 6.27 . q is determined using the bottom width of the chute. K1.7 to 2. Maynord (1996) also gives equations and methods for toe scour estimation and toe scour and end section protection. This erosion is undesirable and can initiate failure of the riprap. The Corps manual EM 1110-2-1601 uses the depth-averaged velocity at a point 20 percent up slope from the toe.20) where: θ φ = = Angle of side slope with horizontal o Angle of repose of the riprap (normally 40 ) The Corps manual ME 1110-2-1601 gives an equation to determine riprap size for steep slopes (2 to 20 percent) where the unit discharge is low.555 q 2 / 3 g1 / 3 (6. (1953) relation in the Corps manual EM 1110-2-1601. toe scour protection. If the riprap design is for a channel bottom. The Corps manual EM 11102-1601 gives the following equation for estimating the side slope riprap design velocity.5 D100.7.74 − 0. D85/ D15 from 1. A typical application is a rock-lined chute.6 in alluvial or man made channels for large discharges for which riprap is designed. The Corps manual 1110-2-1601 gives charts and figures as design guides and guidance for toe scour.S.These factors for determining the value of the safety factor coefficient must be considered and the largest value used as the safety factor. empirical equations or methods. angular rock. Termination of riprap bank protection often leads to the formation of an eddy (wake vortex) that erodes the bank. The relationship is: é Sin 2 θ ù K 1 = ê1 − ú 2 ë Sin φ û 1/ 2 (6. and prototype data.

8 D50 1.2.57 40 0. The recommended gradation for riprap is illustrated in Figure 6.90 60 1.5 D50 0. develops an armor plate.28 20 0.72 50 1.15.6. including sizes up to D50 and larger.6 D50 1.90 D50 D50 D50 D50 D50 D50 D50 D50 D50 6.0 D50 1.2. are transported by the high velocities.70 100 2. some of the finer materials. The computations of the representative grain size Dm for the recommended gradation are given in Table 6.2 D50 1.43 30 0. the size of rock representative of the stability of the riprap is determined by the larger sizes of rock.25 D50 -10 0. The well-graded riprap. A uniformly graded riprap with a median size D50 scours to a greater depth than a well-graded mixture with the same median size. Percent Finer Sieve Diameter DI 0 0.0 D50 0.15 in terms of D50.5. Table 6. Suggested gradation for riprap.5 Riprap Gradation and Thickness The concept of a representative grain size for riprap is simple. Figure 6.35 D50 0. The representative grain size Dm for riprap is larger than the median rock size D50.8 D50 0. on the other hand.10 70 1. Thus. That is. Data for Suggested Gradation.28 .50 90 1. The uniformly distributed riprap scours to a depth at which the velocity is less than that required for the transportation of D50 size rock. leaving a layer of large rock sizes which cannot be transported under the given flow conditions.65 D50 0.

Dm determined by Equation 3. the size which can be obtained economically from the quarry. Riprap consisting of angular stones is more suitable than that consisting of rounded stones. Where available rock size is inadequate.29 .S.23) = 0. and any stone gradation within the limits is acceptable. Lower limit of W 100 stone should not be less than two times the lower limit of W 50 stone. or the size that satisfies layer thickness requirements.22) When the bed material has a log-normal distribution. Considering the practical problems of quarry production.72 D50 and 1. Upper limit of W 15 stone should be less than the upper limit of the filter material.81 D50 respectively. Upper limit of W 100 stone should not exceed five times the lower limit of W 50 stone. the representative size of the bed material based on the weight of the particles is given by Mahmood (1973) as a function of the gradation coefficient G: ì3 ü D m = D 50 exp í (ln G) 2 ý î2 þ (6. The Corps criteria for establishing gradation limits in terms of stone weight (W) for riprap are as follows: • • Lower limit of W 50 stone should not be less than the weight of stone required to withstand the design shear forces. wire enclosed (gabion) riprap can be used. If it is necessary. poor gradations of rock can be employed as riprap provided the proper filter is placed between the riprap and the bank or bed material. of the mixture corresponds to the D65 size of the riprap. or the size that satisfies layer thickness requirements.The rock sizes in the last column in Table 6. a gradation band is usually specified by the U. Dm. Lower limit of W 15 stone should not be less than one-sixteenth the upper limit of W 100 stone. 3 ù é å10 i Di =ê ú ê 10 û ú ë 1/ 3 Dm ≅ 1. Upper limit of W 50 stone should not exceed five times the lower limit of W 50 stone.25 D 50 (6.9. the interstices formed by the larger stones are filled with the smaller sizes in an interlocking fashion. Control of the gradation of the riprap is almost always made by visual inspection. G is For gradation coefficients of 2 and 3.2 are used in the following equation (Stevens 1968) to find the representative grain size Dm. Army Corps of Engineers (1981) rather than a single gradation curve. • • • • 6. With a distributed size range. the size which can be obtained economically from the quarry. preventing formation of open pockets. This effective grain size.

This form of placement is used rarely in modern practice because it is usually more expensive than placement with power machinery. Riprap placement is usually accomplished by dumping directly from trucks. 6. (Figure 6. and its cost is less because a lesser volume of rock per unit area is required.000 lb) or larger piece of steel plate is used to compact the rock into a tight mass and to smooth the revetment surface. The riprap thickness should not be less than 300 mm (12 in. but it is sometimes disregarded for stream banks. 6.5. Broken concrete is used for riprap in many states where rock riprap is unavailable or unusually expensive. whichever is greater. Dumped riprap that is keyed (or plated) by tamping has proved to be effective. Keyed riprap is more stable than loose riprap revetment because of reduced drag on individual stones. rounded stones are nevertheless effective for larger diameters. The blanket should be stabilized at its base with a key trench or apron to prevent the stone from sliding down the bank. Guidelines for placement of keyed riprap have been developed by the Oregon Department of Transportation and distributed by the Federal Highway Administration. The upstream and downstream ends of the blanket should be tied back into the bank to prevent stream currents from unraveling the blanket. The most common method to tie into the bank is to dig a trench at the ends of the blanket.5H. Although they are less desirable than angular stones. usually resulting in a relatively smooth top surface. backhoes. These methods result in segregation of sizes. Draglines. or less than 1. Poorly graded riprap with slab-like rocks requires more work to form a compact protective blanket without large holes or pockets. its angle of repose is higher. rocks can be dumped directly from trucks from the top of the embankment.30 . Hand placed rock riprap is another method of riprap placement. and if it is subject to attack by large floating debris or wave action it should be increased 150 to 300 mm (6 to 12 in. Stones are laid out in more or less definite patterns.• Bulk volume of stone lighter than the W 15 stone should not exceed the volume of voids in the structure without this lighter stone.) for practical placement. Broken concrete riprap has proven to be unsatisfactory if the material is not broken into riprap size particles. and it is more likely to fail than dumped riprap. The depth of a trench should be twice the blanket thickness and the bottom width of the trench three times the thickness.4). In the keying of a riprap. less than the diameter of the upper limit of W 100 stone.16).5 times the diameter of the upper limit W 50 stone.) in diameter have a significantly lower angle of repose than angular stones (Figure 3. Rounded stones less than 150 mm (6 in. the thickness should be increased by 50 percent.). If riprap is placed during construction of the embankment. This criterion is widely followed for abutment fill-slopes. If riprap is placed under water. and other power equipment can also be used to place the riprap. Rock should never be placed by dropping down the slope in a chute or pushed downhill with a bulldozer.818 kg (4.6 Riprap Placement General. With dumped riprap there is a minimum of expensive hand work. a 1. Riprap should not be used at slopes steeper than IV:1.

16. This method is applicable in areas of rapidly eroding banks of medium to large size rivers.Figure 6. Figure 6. Launching aprons can provide similar protection against undermining for channel banks and revetment. A flexible launching apron is placed horizontally on the bed at the foot of the revetment. The size of trench to hold the rock fill depends on expected depths of scour. This method is generally the most economical for cohesionless channel beds where deep scour is expected.31 . the toe trench is undermined and the rock fill slides downward to pave the bank.17. It is advantageous to grade the banks before paving the slope with riprap and placing rock in the toe trench. Rock-fill trench. As the stream bed adjacent to the toe is eroded. Rock-fill trenches are structures used to protect banks from caving caused by erosion at the toe. Rock-fill Trenches and Launching Aprons. The rock-fill trench need not be at the toe of the bank. A trench is excavated along the toe of the bank and filled with riprap as shown in Figure 6. An alternative method is to excavate a trench above the water line along the top of the river bank and backfill with rocks. so that when scour occurs the materials will settle and cover the side of the scour hole on a natural slope. The slope should be at such an angle that the saturated bank is stable while the river stage is falling.17. Then as the bank erodes toward the trench. Tie-in trench to prevent riprap blanket from unraveling (after Keown 1983). the rocks in the trench slide down and pave the bank. Materials used for launching aprons include 6.

beyond which additional erosion is to be prevented.stone riprap. Windrow revetment is an erosion control technique (Figure 6. but little field confirmation seems to have been reported. (3) Studies indicated that varying the bank height did not significantly affect the final revetment. and the excavation then refilled. there should be sufficient stone to cover the final conical surface of the scoured slope. Then as the bank is scoured.32 . with an irregular appearance due to intermittent lateral erosion in the windrow location. Launching aprons do not perform well on cohesive channel beds where scour occurs in the form of slumps with steep slip faces. The technique consists of burying or piling a sufficient supply of erosion-resistant material in a windrow below or on the existing land surface along the bank. Model tests have indicated that these slopes are realistic for sand beds. thus giving protection against further undercutting. At the nose of a guide bank or spur. A triangular shape was found to be the least desirable. or placement along stretches of presently eroding. The following observations and conclusions were obtained from model investigations on windrow revetments. Windrow Revetment. and 6. gabions. then permitting the area between the natural riverbank and the windrow to erode though natural processes until the erosion reaches and undercuts the supply of rock. A variation of these methods of toe protection is to pile the rocks in a "windrow" along the bank line instead of excavating a trench. articulated concrete matting. the design of stone aprons should be based on the stone launching to a slope of up to lV:2H. (2) Various windrow shapes were investigated in the model investigations. Piers should not be located within the launching apron slope unless it is unavoidable. In such cases. Studies showed that the high banks had a tendency for large segments of the bank to break loose and rotate slightly. (1) The "application rate" is the weight of stone applied per meter (foot) of bank line. it falls onto the eroding area.18) consisting of the depositing of a fixed amount of erosion-resistant material (riprap) landward from the existing bank line at a predetermined location. The slight rotation of the high bank segment probably induced a tendency for ragged alignment. In cohesionless channel beds. Stone riprap is most commonly used. The treatment particularly lends itself to the protection of adjacent wooded areas. and wire mesh mattresses filled with stone. The amount of stone in the windrow indicates the degree to which lateral erosion will be permitted to occur. whereas the low banks simply "melted" or sloughed into the stream.25 times the size of the largest stones in the specified grading. and eventually halting further landward movement. The volume of stone should be sufficient to cover the final scoured slope to a thickness of 1. the rocks in the windrow drop down to pave the bank. As the rock supply is undercut. The second best windrow shape was found to be a trapezoidal shape. concrete blocks. The resulting bank line remains in a near natural state. however. This shape provides a steady supply of stone to produce a uniform blanket of stone on the eroding bank line. This type of windrow is most easily placed in an excavated trench of the desired width. bank revetment should be continued down to the expected worst scour level. high banks tended to produce a nonuniform revetment alignment. and a rectangular cross section was the best windrow configuration. Stone sizes for launching should be the same as for slope revetment. irregular bank line.

FHWA HEC-15. Filters must meet two basic requirements: stability and permeability.18. 6.33 . It was found that large stone sizes will require more material than smaller stone sizes to produce the same relative thickness. The filter material must be fine enough to prevent the base material from escaping through the filter. An important design parameter is the ratio of the relative thickness of the final revetment to the stone diameter. (4) The velocity and characteristics of the stream dictate the size of stone that must be used to form a windrow revetment. The size of stone used in the windrow was not significant as long as it was large enough to resist being transported by the stream. Blodgett and McConaughy (1985) concluded that the procedures based on velocity as a means of estimating stresses on the boundary provide the most reliable and consistent results. the greater the velocity. Since a filter cannot be used. In general. a well-graded stone is important to ensure that the revetment does not fail from leaching of the underlying bank material.7 provides general specifications for granular and geosynthetic filters. Filters. Filters are used under riprap to allow water to drain easily from the bank without carrying out soil particles. HEC-11 (Brown and Clyde 1989) contains more detailed design guidance.7 Riprap Failure Modes In a preliminary evaluation of various riprap design techniques. 6.S. definition sketch (after U. The stream velocity was found to have strong influence on the ultimate side slope of the revetment. Section 6. The following procedures were investigated: The 1967 version of FHWA HEC-11. USACE (EM-1601). Windrow revetment. the steeper the side slope of the final revetment.Figure 6. It was determined that the initial bank slope was on the average approximately 15 percent steeper than the final revetment slope. but it must be more permeable than the base material.5. Army Corps of Engineers 1981).

19b). but the geometry of the damaged riprap is similar in shape to initial stages of failure caused by particle erosion. Classic riprap failure modes are identified as follows: (1) particle erosion. and Oregon Department of Transportation. which is usually caused by excess pore pressure that reduces friction along a fault line in the base material.19. particle erosion. (2) individual stones removed by impact or abrasion. (2) presence of excess hydrostatic (pore) pressure. Particle erosion is the most commonly considered erosion mechanism (Figure 6. or some other mechanism which causes displacement of toe material.19c) is the mass movement of material along an internal slip surface within the riprap blanket. impact. freeze/thaw action.Caltrans Bank and Shore Manual. Probable causes of translational slides are as follows: (1) bank side slope too steep. Modified slump failure of riprap (Figure 6. A translational slide is a failure of riprap caused by the downslope movement of a mass of stones. This type of failure is similar in many respects to the translational slide. Probable causes of modified slump are: (1) bank side slope is so steep that the riprap is resting very near the angle of repose. the underlying material supporting the riprap does not fail. The cause of slump failures is related to shear failure of the underlying base material that supports the riprap. or toe erosion. or some other cause. Particle erosion can be initiated by abrasion. and (4) slump. The primary feature of a slump failure is the localized displacement of base material along a slip surface.19d). Simons and Senturk. It has been suggested that the presence of a filter blanket may provide a potential failure plane for translational slides. and (3) loss of foundation support at the toe of the riprap blanket caused by erosion of the lower part of the riprap blanket. Particle erosion occurs when individual particles are dislodged by the hydraulic forces generated by the flowing water.19a). This type of riprap failure is usually initiated when the channel bed scours and undermines the toe of the riprap blanket. (2) translational slide: (3) modified slump. A major shortcoming of all present design techniques is their assumption that failures of riprap revetment are due only to particle erosion. with the fault line on a horizontal plane (Figure 6. and (2) material critical to the support of upslope riprap is dislodged by settlement of the submerged riprap. Probable causes of particle erosion include: (1) stone size not large enough. Procedures for the design of riprap protection need to consider all the various causes of failures. and (2) side slopes too steep and gravitational forces exceeding the inertia forces of the riprap and base material along a friction plane. Slump failure is a rotational-gravitational movement of material along a surface of rupture that has a concave upward curve (Figure 6. ice.34 . 6. eddy action/reverse flow. and (4) gradation of riprap too uniform. local flow acceleration. and any imbalance or movement of individual stones creates a situation of instability for other stones in the blanket. Any other mechanism which would cause the shear resistance along the interface between the riprap blanket and base material to be reduced to less than the gravitational force could also cause a translational slide. impingement of flowing water. (3) side slope of the bank so steep that the angle of repose of the riprap material is easily exceeded. These modes of failure are illustrated in Figure 6. The initial phases of a translational slide are indicated by cracks in the upper part of the riprap bank that extend parallel to the channel. this could be caused by particle erosion of the toe material. Probable causes of slump failures are: (1) nonhomogeneous base material with layers of impermeable material that act as a fault line when subject to excess pore pressure. abrasion.

Figure 6. 6.35 .19. Riprap failure models (after Blodgett and McConaughy 1985).

tree root systems do not always penetrate to the toe of the bank. Vegetation is generally divided into two broad categories: grasses and woody plants (trees and shrubs). the weight of the tree and its root mass may cause a bank failure. Second. flatten riprap slopes. Woody plants offer greater protection against erosion because of their more extensive root systems. branches and foliage provide resistance to the streamflow.1 Bioengineering Erosion Control Vegetation is probably the most natural method for protecting streambanks because it is relatively easy to establish and maintain. The principal causes of failure and methods of mitigation are given in Table 6.3. under some conditions the weight of the plant will offset the advantage of the root system. vegetation takes water from the soil providing additional capacity for infiltration and may improve bank stability by water withdrawal. the exposed stalks.Because of the general effectiveness of dumped riprap.36 . If the toe becomes eroded. is visually attractive and environmentally more desirable. vegetation can effectively protect a bank in two ways. baskets. redirect flow Provide a volume of reserve riprap at the revetment toe Reduce the riprap slope angle Follow gradation specifications 6. Table 6. This section provides an overview of several of these methods. Above the waterline. including: vegetation. and blocks. Below a stream's waterline. vegetation prevents surface erosion by absorbing the impact of falling raindrops and reducing the velocity of overbank drainage flow and rainfall runoff.6. 2001) and HEC-11 (Brown and Clyde 1989) provide detailed design guidance for several of these methods (see Section 6. 6. The grasses are less costly to plant on an eroding bank above the toe and require a shorter period of time to become established. the root system helps to hold the soil together and increases overall bank stability by forming a binding network.3. 6. mattresses. however. Causes of Riprap Failure and Solutions (Brice et al. stems. causing the flow to lose energy by deforming the plants rather than by removing soil particles. On very high banks. HEC-23 (Lagasse et al.6 BANK PROTECTION OTHER THAN RIPRAP There are many methods other than riprap that can be used for bank protection. a more detailed analysis of the relatively small number of cases in which it failed has been presented by Brice and Blodgett (1978). Further. First.3.4). 1978) Cause Inadequate size of riprap Impingement of current directly upon riprap rather than having flow parallel to riprap Channel degradation Internal slope failure (slump) Riprap with high percentage of fines causes washing out of the fines Solution Larger riprap Heavier stones.

Although most of the literature dealing with biotechnical engineering on rivers is 6. bulrushes (Scripus). however. but the risk associated with using just vegetation is considered too high. gates should be placed in the fence at locations where the cattle will do the least amount of damage to the planted bank.2 Bioengineering Countermeasures The past few decades have seen increasing use of vegetation as a bank stabilizer. The matting is generally installed by hand and secured to the bank where plantings have been made to prevent erosion. geomorphic. reeds (Phragmites).The major factor affecting species selection is the length of time required for the plant to become established on the slope. However. and mannagrass (Glyceria) are helpful in inducing deposition and reducing velocities in shallow water or wet areas at the bank toe and in protecting the bank in some locations. or by mechanical broadcasting of mulches consisting of seed. Bioengineering erosion control is not suitable where flow velocities exceed the strength of the bank material or where pore water pressure causes failures in the lower bank. are sufficiently dense to promote deposition of sediment. and become established easily. The term bioengineering and is generally used to describe stream bank erosion countermeasures and bank stabilization methods that incorporate vegetation. and construction factors. grasses are generally not effective as a protective measure. where erosive forces are high. cattails (Typha). In contrast. reedgrass (Calamagrostis). Along the lower bank. can withstand inundation. There is research being conducted in these fields. Water-tolerant grasses such as canarygrass (Phalaris). Stabilization of eroding stream banks using vegetative countermeasures has proven effective in many documented cases in Europe and the United States. bioengineering maybe suitable where some sort of engineered structural solution is required. Several commercial manufacturers market economical erosion control matting that will hold the seed and soil in place until new vegetation can become established. knotweed and smartweed (Plygonum). and other organic mixtures. but these techniques have generally not been tested specifically as a countermeasure to protect bridges in the river environment. then a fence should be placed along the top of bank. Willows (Salix) are among the most effective woody plants in protecting low banks because they are resilient. sodding. additionally.37 . this group of countermeasures is not as well accepted as the classical engineering approaches to bridge stability. vegetative. geotechnical. fertilizer. It has been used primarily in stream restoration and rehabilitation projects and can be applied independently or in combination with structural countermeasures. 6. and fescue (Festuca) are effective in prevent erosion on upper banks which are inundated from time to time and are primarily subject to erosion due to rainfall. Design of bioengineered countermeasures to minimize rates of stream bank erosion requires accounting for hydrologic. sprigging.6. crossings should be fenced. Nonetheless. rushes (Juncus). the use of bioengineering with respect to scour and stream instability at highway bridges is a relatively new field. hydraulic. and minor wave action. cordgrass (Spartina). If livestock require access to the stream for watering or crossing. Grass can be planted by hand seeding. overland flow.

Special wire baskets and mattresses are manufactured and sold throughout the United States.20.). The term mattress implies a thickness no greater than 300 mm (12 in. It should be noted that when rock-and-wire mattresses are used in streams transporting cobble and rocks. The individual wire units are called baskets if the thickness is greater than 300 mm (12 in. Figure 6. They should be linked together to prevent separation as subsidence takes place. studies. rocks of cobble sizes may be placed in wire mesh mats made of galvanized fencing and placed along the bank forming a mattress. HEC23 (Lagasse et al. the extended mattress drops into the scour hole.). As the bed along the toe is scoured. HEC-23 (Lagasse et al. 2001) contains specifications developed by the New Mexico State Highway and Transportation Department for wire enclosed riprap mattress.associated with stream bank stabilization relative to channel restoration and rehabilitation projects. Corrosion of the wire mesh and vandalism may also be a problem. 6. Toe protection is offered by either extending the mattresses on to the channel bed as shown in Figure 6.20 or embedding the mattress to some predetermined scour depth. 6.38 . 2001) provides an overview of bioengineering approaches. and handbooks containing design guidelines. These are flexible and can conform to scour holes which threaten the stability of the banks. and reference to recent reports. a summary of design considerations.3 Rock-and-Wire Mattresses When adequate riprap sizes are not available. it is also generally applicable to bank stabilization associated with bridge crossings. which will destroy the intended protection along the base of the bank. Mattresses and baskets can be made up in large sizes in the field.6. the wires of the basket can be cut by abrasion rather rapidly. Rock and wire mattress.

this type of sack revetment can often be placed on an eroding streambank at a lesser cost than riprap. However.3 x 6. to prevent leaching of base material and undermining of the baskets. plastics. riprap can be placed at the toe to prevent undermining of the bank slope).4 to 4. One successful solution has been to bend pipe or rebar into the shape of a staple and then drive it through the mesh into the bank. using noncorrosive wire. Gabions are used primarily for revetment-type structures.) in diameter. A filter blanket or synthetic filter fabric is used. 6. The filled sacks should be placed in horizontal rows like common house brick beginning at an elevation below any toe scour (alternatively.21).5 ft) and are set on a graded bank for revetments.6 m/s (8 to 15 ft/s). depending on the manufacturer.) have been used to protect streambanks in areas where riprap of suitable size and quality is not available at a reasonable cost. After the sacks have been placed on the bank. 6. they can be hosed down for a quick set or the sand-cement mixture can be allowed to set up naturally through rainfall. a bond will form between the sacks and prevent free drainage.39 . The major problem with this approach is keeping the mesh in place. The installation of weepholes will allow drainage of groundwater from behind the revetment thus helping to prevent pressure buildup that could cause revetment failure. In recent years commercially manufactured sacks (burlap. Detailed design guidelines for commercially available articulating grout filled mattresses are provided in HEC-23 (Lagasse et al. If cement leaches through the sack material. The baskets are normally 0. The major drawback is that a rock and wire mattress generally costs more to place than a comparable riprap blanket.5 Sacks Burlap sacks filled with soil or sand-cement mixtures have long been used for emergency work along levees and streambanks during floods (Figure 6. They are supplied at a job site folded flat and are assembled manually. Where flow velocities are such that small stone would not be stable if used in a riprap blanket. paper. Limiting recommended maximum velocity for use of gabions ranges from 2. etc. Although most types of sacks are easily damaged and will eventually deteriorate. the wire boxes provide an effective restraint. Sand-cement sack revetment construction is not economically competitive in areas where good stone is available. HEC-11 (Brown and Clyde 1989) provide design guidelines for gabion revetment. For this reason weepholes should be included in the revetment design.6 x 3.5 m deep by 1 m by 2 m (1. The slope steepness of the completed revetment should be no more than 1:1. The baskets are commercially available in a range of standard sizes and are made of heavy galvanized wire (coated when used in a corrosive environment). 6.6. seepage or condensation.6. if quality riprap must be transported over long distances. The successive rows should be stepped back approximately 1/2-bag width to a height on the bank above which no protection is needed. but have also been used for dikes and sills. 2001).The most economical combination of rock and wire for streambank protection is simply laying wire mesh over stone. the sacks should be filled with a mixture of 15 percent cement (minimum) and 85 percent dry sand (by weight). those sacks filled with sand-cement mixtures can provide long-term protection if the mixture has set up properly. Gabions act as a large heavy porous mass having some flexibility. If a permanent revetment is to be constructed. where required. usually less than 200 mm (8 in.4 Gabions Gabions are patented rectangular wire boxes (or baskets) filled with relatively small-size stone.

hand placement is frequently used when mechanized apparatus is not available.21. Solid blocks should not be used because the bank may not be able to drain freely and failure could occur. It is particularly difficult to make a continuous mattress of uniform sized blocks to fit sharp curves. Small precast concrete blocks held together by steel rods or cables can be used to form a flexible mat as shown in Figure 6. or costs need to be minimized.6. and aggregate or can be obtained from commercial sources. After the blocks have been placed. Although specialized equipment can be used to install large sections of blocks.6 Articulated Concrete Block Systems (ACBs) Precast cellular blocks can be manufactured using locally available sand. 6.22. Typical sand-cement bag revetment (after Keown 1983).Figure 6. The open spacing between blocks permits removal of bank material unless a filter blanket of gravel or geosynthetic material is placed underneath. the spaces between blocks may be filled with earth and vegetation can be established. access to the bank is limited.40 . Fabric or a gravel blanket can be used as a filter under the blocks if there is any danger that the bank soil will be eroded through the block openings by streamflow or seepage. The sizes of blocks may vary to suit the contour of the bank. Cellular blocks are cast with openings to provide for drainage and to allow vegetation to grow through the blocks thus permitting the root structure to strengthen the bank. Design guidelines for ACBs are provided in HEC-23 (Lagasse et al. the revetment has sufficient flexibility to conform to minor changes in bank shape. 2001) and HEC-11 (Brown and Clyde 1989). 6. cement. For embankments that are subjected only to occasional flood flows.

Figure 6. The top. 6. • • The tires must be banded together.23. Both methods appear to have some potential as an economical approach to protect a streambank. but environmental issues and aesthetics must be considered. alternatively.41 .7 Used Tires Tires have been placed both as a mattress and stacked back against the bank.Figure 6. two precautions should be considered to ensure that the mattress will stay in place. cables running the length and width of the mattress can be woven through the tires.6.22 Articulated concrete block system. toe and the upstream and downstream ends of the mattress must be tied to the bank (Figure 6. If scour is anticipated. riprap should be placed at the toe of the mattress for additional protection. 6. During construction of a tire mattress on an eroding bank.23). Used tire mattress (after Keown 1983).

or burned in the tire sidewalls to prevent flotation. The tires can be packed with stone or rubble.42 . 200 sieve. other considerations can further improve the chances that the revetment will provide long-term bank protection. 6. silt and clay (material passing the No.6. Possible species for use are discussed later in this text. but not more than 35 percent. A stairstep construction is recommended on channel embankments with relatively steep slopes. 2001) provides design guidelines and specifications for soil cement revetments. Each tire should overlap the two tires under it. For use in soil-cement. a sheepsfoot roller should be used on the last layer to provide an interlock for the next layer. Presorting the tires by size may help to fit them together. the upstream and downstream ends of the revetment should be tied into the bank so that there is no flow behind the revetment. If willows are not readily available.While the precautions listed above are essential for successful construction of a stable mattress. the material has sufficient strength to serve as a roadway along the embankment." Portland Cement Association Publication IS052W and numerous other PCA publications. After completion.24 shows a detail of a typical soil-cement construction for bank protection. If the mattress effectively controls the streambank erosion and remains intact. Special care should be exercised to prevent raw soil seams between successive layers of soil-cement. Figure 6. 200 sieve). In addition. drilled.8 Soil Cement In areas where riprap is scarce. Once established the root system will further strengthen the bank and obscure the unsightly mattress. soils should be easily pulverized and contain at least five percent. Finer textured soils usually are difficult to pulverize and require more cement as do 100 percent granular soils which have no material passing the No. If willows have not been planted. The revetment should be started at the toe of the bank and stepped back 150 to 300 mm (6 to 12 in. Soil cement can be placed and compacted on slopes as steep as one horizontal to one vertical. The stacked tires should be packed tightly with stone or rubble. 6. Willows can be planted inside the tires preferably at the beginning of the growing season. the bank face should be shaped so that the tires can be laid in horizontal rows on a geosynthetic filter material. use of in-place soil can sometimes be combined with cement to provide a practical alternative. Placement of small quantities of soil-cement for each layer 150 mm (6-inch) layers can progress more rapidly than a large quantity of fill material. Any space behind the tires should be filled with free-draining soil so that the soil mass will not become saturated and cause the revetment to fail. Procedures for constructing soil-cement slope protection by the stairstep method can be found in "Suggested Specifications for Soil-Cement Slope Protection for Embankments (Central-Plant Mixing Method). volunteer vegetation will may become established in areas with a temperate climate. sediment may gradually cover the revetment. Earth screw anchors (or some other type of anchor) fastened to the mattress can be placed in the bank at various points on the face of the revetment. The completed soil-cement installation must be protected from drying out for a seven day hydration period. Prior to constructing a stacked-tire revetment. • • • • • Holes can be cut. other species should be planted. If uncompleted embankments are left at the end of the day.) per row. HEC-23 (Lagasse et al.

This method has the advantage of low cost. Sizes of gravel in the filter blanket should be from 5 mm (3/l6 in. special precautions are advisable.75 mm) sieve. Also.). there are three major disadvantages: impermeability.7 FILTERS Filters are used under riprap to allow water to drain easily from the bank without carrying out soil particles.). because a sand-cement blanket is relatively brittle.Figure 6. low strength.43 . and susceptibility to temperature variations. A layer or blanket of well-graded gravel should be placed over the embankment or riverbank prior to riprap placement.8 to 2. There is no standard filter that can be used in all cases. The depth of the bank protection should be sufficient to protect the installation from the anticipated total scour. 6. or livestock) can be sustained without cracking the thin protective veneer. see HEC-11 (Brown and Clyde 1989) for additional guidelines on filter design. A soil cement blanket with 8 to 15 percent cement may be an economical and effective streambank protection method for use in areas where vegetation is difficult to establish and the bank material is predominately sand. 4 (4. Two types of filters are commonly used: granular (gravel) filters and geosynthetic filters. When velocities exceed 1. If the bank behind the blanket becomes saturated and cannot drain. Filters must meet two basic requirements: stability and permeability. Thickness of the filter may vary depending upon the riprap thickness but should not be less than 152 to 228 mm (6 to 9 in. pedestrian. In northern climates the blanket can break up during freeze-thaw cycles. The aggregates in this case should contain at least 30 percent gravel particles retained on a No. Typical soil-cement bank protection. It should be emphasized that soil-cement provides a rigid bank protection.24. Filters that are one-half the thickness of the riprap are quite satisfactory. The filter material must be fine enough to prevent the base material from escaping through the filter. very little if any traffic (vehicular. The sand can be mixed with cement by hand or mechanically to a depth of at least 4 inches.) to an upper limit depending on the gradation of the riprap with maximum sizes of about 76 to 89 mm (3 to 3-l/2 in. Granular Filters. failure may occur.4 m/s (6 to 8 ft/s) and the flow carries sufficient bed load to be abrasive. The mixture should then be wet down and allowed to set up. However. but it must be more permeable than the base material. Suggested specifications for gradation are as follows: 6.

" Numerous geotextiles are on the market. overlapped.000 lbs) have been placed on synthetic filters with no apparent damage. If the filter is designed for protection against the upward flow of water. and more than one layer of filter (a graded filter) may be needed. When the base material is very fine. approach embankments. with a wide variation in size and number of openings and in strength and durability. these requirements are not applicable. See Holtz et al. care must be taken not to puncture the material during construction. so that the filter fabric is not punctured. however. Care is also required in joining adjacent sections of filter fabric together. levees." where tailwater may be present against the downstream embankment slope but at an elevation low enough such that interference with flow over the crest does not occur. The sides and toe of the filter fabric must be sealed or trenched so that base material does not leach out around the filter fabric.1 Introduction Floodwaters which exceed the crest elevation of roadways.24) ( 2) 5 < (6. Detailed information on the use of geosynthetic filters can be found in Holtz et al. If the paving materials is dumped or cast stone. the required filter material may also be quite fine." where tailwater is higher than the embankment crest and presents a backwater condition sufficient to affect the discharge over the crest.4 mm.25) (3 ) D15 (Filter ) <5 D 85 (Base ) (6. and the stability criterion is that the D15 size of the filter cannot exceed 0.8 OVERTOPPING FLOW ON EMBANKMENTS 6. the graded filter is constructed so that each layer is coarser than the one beneath (a "reverse" or "inverted" filter). Geotextiles which provide opening areas of 25 to 30 percent are desirable to minimize the possibility of clogging and to reduce head loss. and welted seams are used. Stones weighing as much as 1. each layer must satisfy the stability and permeability requirements relative to the underlying layer. Geosynthetic materials are also used as filters. and (2) "free flow.360 kg (3. Overtopping flow can be broadly characterized into two categories according to the level of tailwater on the downstream side of the structure: (1) "submerged flow. replacing a component of a graded filter. 6. gravel can sometimes be placed directly on the fabric. If the filter fabric is placed on top of the base material. 1995 (FHWA HI-95-038) who define permeable geosynthetics as "geotextiles. 1995 for AASHTO specifications and construction guidelines.8. such as fat or lean clay. it is desirable to place a protective blanket of sand or gravel on the filter. sewn. the size and drop of the rock should be limited. 6. eliminating the need for filter sand. When geotextiles are used. Geosynthetic Filters. or to take care in placing the rock.44 . and similar earth embankment structures result in a hydraulic condition referred to as overtopping flow.(1) D 50 (Filter ) < 40 D 50 (Base ) D15 (Filter ) < 40 D15 (Base ) (6.26) If the base material is a fine-grained cohesive soil. In such a case. If a protective covering is not used.

and the prevention of erosion under varying conditions of soil types. the mechanics of overflow erosion. and erosive energy in this flow zone is low. 6.Since the mid–1980s. Hydraulic flow zones on an embankment during overtopping flow (Powledge et al. 1989b). The ASCE Task Committee on the Mechanics of Overflow Erosion on Embankments (Powledge et al.45 . the erosive energy increases significantly as the flow accelerates down the slope.8. The first type of flow condition is known as free flow. Field studies and laboratory investigations conducted both at full scale and at model scales during the last two decades have contributed to the understanding of analysis and design issues associated with overtopping flow. including the hydraulic conditions of flow. and embankment geometry. Kindsvater (1964) was one of the first to classify these conditions. hydraulic conditions. Figure 6. In the case where there is minimal tailwater during an overtopping event. Figure 6.25. The energy slope increases as well. 1989b) summarizes more than 30 dam. Estimating Discharge. an increase in velocity creates an increase in local shear stress. In Zone 1. In this zone the total energy is essentially equal to the elevation of the low-velocity pool upstream of the embankment. Zone 2 consists of a transition zone near the downstream shoulder where the flow transitions from subcritical to supercritical. While the total energy remains similar to that in Zone 1.2 Hydraulics of Overtopping Flow Flow Zones. extensive research has been conducted on overtopping flow over embankments and on the effectiveness of alternative methods for protecting the downstream slope of embankments against erosion during overtopping events. Many of these research activities are also summarized by the Task Committee (Powledge 1989a). and the erosive energy begins to build. A number of different aspects of overtopping flow have been investigated. Zone 1 is characterized by subcritical flow over the dam crest. Zone 3 consists of supercritical flow on the downstream slope of the embankment. In this zone. levee and roadway embankments that experienced overtopping flows during the 1980s. flow over an embankment structure transitions through three zones. the embankment will experience erosion only when its crest is comprised of highly erodible material. Flow over an embankment can exhibit two types of behavior. 1989b).25 illustrates these flow regimes (Powledge et al. causing high erosion potential. and is characterized by tailwater conditions that are low 6.

51 for SI units or 3. The term plunging flow is used when the high-velocity jet plunges under the tailwater surface.26 illustrates the variables needed to describe overtopping flow (Kindsvater 1964). creating a hydraulic jump. This program was used to develop 6. C ≈ 0. Analysis of these studies developed embankment erosion equations and a computer simulation program called EMBANK (Chen and Anderson 1987). Colorado. Figure 6. Free flow is further subdivided into either free-plunging or free-surface flow. m (ft) 2 2 H1 = The second flow behavior is known as submerged flow. Definition sketch of variables involved in overtopping flow (Kindsvater 1964). This facility was used to carry out a two-phase research project. and prototype-scale embankment tests involving different types of protection systems. the equation becomes æC ö q = C H3/2 ç s ÷ è C ø (6. For the case of submerged flow. Figure 6. field studies. FHWA initiated research for which a full-scale testing facility was constructed in Fort Collins. the tailwater is high enough to affect the discharge over the embankment. In 1983. Hydraulic Analysis.enough such that backwater conditions on the crest do not occur. Surface flow occurs when the jet separates from the downstream slope of the embankment and "rides" over the tailwater surface.26.0 for English units) Total head above the embankment crest. In this case. Phase I consisted of an extensive literature review.46 . The generally accepted equation for discharge over an embankment is 3/2 q = C H1 (6. m /s (ft /s) Experimentally determined coefficient (for level-crested embankments.27) where: q C = = Discharge per unit width.28) where Cs is the submergence coefficient and the term Cs/C is dimensionless.

27. If uniform flow is assumed.design charts that could be applied to estimate embankment erosion caused by a range of overtopping depths and tailwater conditions. ρ is the unit mass of water. Using the principle of the conservation of momentum. it is often directly related to the bed shear stress that is applied as the flow accelerates over and down the embankment. q is the volumetric discharge per unit width. Figure 6. Manning’s equation can be used to estimate the depth of flow. τo = é 1 öù γ 2 2æ 1 ÷ (d1 + d 2 ) sinθ + 1 ê γ d1 − d2 − 2 cos θ − ρq ç ç ÷ú 2 Lë ê2 ú è d 2 d 2 øû ( ) (6. the following equation was developed for non-uniform flow on the downstream embankment slope. the assumption is made that the friction slope Sf is equal to the bed slope S0.47 . When a bare soil embankment fails.27. unless the combination of slope length and unit discharge result in conditions which approach uniform flow. 6. and shear stress calculated by τo = γ d S f (6. This assumption usually yields a conservatively high estimate of bed shear stress.29) where γ is the unit weight of water. and the remaining parameters are presented in Figure 6.30) It should be noted that when using Manning’s equation. Clopper and Chen (1988) describe a computational approach to calculating the bed shear stress. Definition sketch for application of the momentum equation for embankment overtopping flow.

10 percent of the base value be used to calculate the range of anticipated flow conditions. The downstream migration of the erosion is caused by the turbulence associated with the hydraulic jump. erosion typically begins at the downstream shoulder. This is the condition often experienced by roadways and approach embankments.28 (Chen and Anderson 1987) shows the progression of this type of failure at times t1. As flow accelerates over the embankment. This nick point will begin to migrate upstream due to the high velocities and erosion will begin to move downstream.8. When flow overtops an embankment. 6. a surging hydraulic jump is formed which causes a nick point between the shoulder and downstream slope. This is because a balance of forces can develop once the initial embankment erosion has produced a geometry that achieves equilibrium between hydraulic forces and the erosion resistance of the remaining embankment material. depending on the tailwater conditions and embankment geometry and soil type. It should be noted that. 6. and the maximum value of Manning’s n is used to determine peak anticipated flow depth and shear stress. locally high velocities and shear stresses will create strong erosion forces. that are too great for the soil of the embankment to withstand.3 Mechanics of Overflow Erosion Erosion Processes. Figure 6.28. When the overtopping flow is submerged. t2. even events of long duration may not ultimately result in a full breach of the embankment section.48 . Typical embankment erosion pattern with submerged flow (Chen and Anderson 1987). it is recommended that a range of Manning’s n values of approximately +/. The minimum value of Manning’s n is used to determine the peak anticipated velocity. typically at the downstream shoulder and on the embankment slope.In practice. Figure 6. and t3. There are two primary processes of erosion that occur during an overtopping event.

Chen and Anderson (1987) present the following equation (in English units) for calculating the shear stress in the immediate vicinity of the erosion location.29. For an embankment undergoing erosion. Figure 6.31) where: E τ τc K&a = = = = Detachment rate per unit area Local effective shear stress based on hydraulic conditions Critical shear stress of the soil Empirical coefficients dependent on soil properties Local Shear Stress Calculation. Figure 6. Most of these equations use the effective shear stress and/or velocity along with some measure of resistance of the embankment material to estimate the rate of soil loss. the flow will accelerate down the slope with high velocity and shear stress associated with supercritical flow. With low tailwater.τ c ) a (6. another parameter of importance is local shear stress.29 (Chen and Anderson 1987) illustrates this progression. Erosion progresses in the upslope and upstream direction through the embankment. Erosion typically initiates near the toe of the embankment.49 . These equations were assessed (Chen and Anderson 1987) by relating the erosion rates measured during tests to those calculated by the various equations. The erodibility rate relation developed by the Agricultural Research Laboratory was found to be the most applicable to embankment overtopping flow conditions based on these studies. whether or not a hydraulic jump is present. Erodibility Rate Equation. 6. This general form of the equation is: E = K (τ . this type of erosion pattern typically will result in a full breach of the embankment section.The second general erosion pattern results from the case of free flow. Given sufficient duration. Typical embankment erosion pattern with free flow (Chen and Anderson 1987). Many analytical equations have been presented to estimate the rate of erosion under conditions of overtopping flow.

31 shows a typical full-scale test of an articulated concrete block system in progress under steep-slope. The permissible shear stress from the controlled testing program is then extended to various conditions of bed slope and side slope. and are empirical in nature. Both types of methods are described in this section.33) and V is the local velocity.32) where: f= 116 n 2 d1/3 (6. Note that the permissible velocity of concrete-based systems are independent of duration. described previously in this section. Figure 6. procedures based on permissible velocity are generally derived from extensive testing and field experience using a particular material under a variety of flow conditions. Hewlett et al (1987) present permissible velocities for reinforced grass channels. f is the Darcy-Weisbach coefficient. Permissible Velocity Approach. Typically. a discrete-particle approach to the stability analysis is developed using the force. This form of the shear stress equation is utilized in the program EMBANK. Laboratory or field tests are recommended in order to develop the calibration parameters needed to fully describe the performance of the system. e.4 Erosion Protection in Overtopping Flow Materials or manufactured systems designed to protect against overtopping erosion can be selected and designed using methods based on permissible velocity. Permissible Shear Stress Approach.30 as a function of duration of the overtopping event. These approachs are presented in Hydraulic Engineering Circular No. and grout-filled fabric mats. 23 (Lagasse et al. as shown in the figure. Permissible shear stress based approaches are usually related to force-balance (sliding) or moment-balance (overturning) representations of the stability paradigm. 6. or both. Based on a battery of tests of vegetated trapezoidal channels using various types of reinforcement materials on steep waterways. The use of the local shear stress equation can assist in accommodating the irregular pattern of erosion and its nonuniform progression through an embankment section at various times during an overtopping event.τ= 1 f ρ V2 8 (6. and ρ is the water density. permissible shear stress. n is Manning's resistance coefficient. concrete armor units. Because velocity in and of itself is not a force. and are typically developed using a force-balance or moment-balance concept.8. Procedures utilizing shear stress as a fundamental variable tend to be more theoretical in nature.50 . These curves are presented in Figure 6.or moment-balance equations for the particular system under consideration. articulating concrete blocks. 2001) for selected erosion control systems. and d is the local flow depth.g. high-velocity flow.. 6.

All reinforced grass values assume well established. 3. Values are based on available experience and information at the date of this report. ease of installation and management.8 inch) Installed within 20 mm (0. 4. 2. 1994): 1. 6. 7. Minimum superficial mass 135 kg/m (28 lb/ft ) Minimum nominal thickness 20 mm (0. These systems are described in detail in a summary report issued by the ASCE Task Committee on Overtopping Protection (Oswalt et al. Reinforced Vegetation Roller Compacted Concrete Soil Cement Cast in Place Concrete Articulating Concrete Blocks Geotextile Riprap Gabions 2 2 6.Figure 6. susceptibility to vandalism) must be considered in choice of reinforcement.8. or in conjunction with a surface mesh These graphs should only be used for erosion resistance to unidirectional flow. Permissible velocities of various protection materials as a function of flow duration (Hewlett et al. 8. 3. 5. 6. 1987). Listed below are protection systems which have been the subject of various investigation over the last several decades. 6. 4.5 Overtopping Protection Systems A number of material types and manufactured systems have been identified for use in minimizing or preventing erosion of embankments subjected to overtopping flow. 5.8 inch) of soil surface.30. 2.51 . Other criteria (such as short-term protection. good grass cover. Notes to Figure: 1.

31.Figure 6.52 . Full-scale test of an embankment overtopping protection system under steep-slope. 6. high-velocity flow conditions.

6. While general categories of impacts may be stated. reference to Julien (1995) is suggested.1 Environmental Impacts Impacts of streambank protection projects are dependent on project location and regional characteristics. and stream hydraulics. for example. and other water quality parameters. Powledge. For example. however.9 ENVIRONMENTAL CONSIDERATIONS Streambank protection projects should be planned. For application of the Factor of Safety method (Section 6. but changes in stream hydraulics caused by the project can result in adverse impacts throughout the floodplain ecosystem.53 . sediment-carrying capacity of the stream. (1989a. site-specific and regional considerations and individual project features are critical in determining the magnitude and type of environmental impact. Categories of environmental impacts associated with streambank protection projects include aesthetic. Environmental quality should address preservation or restoration of environmental resources within the project boundaries and avoidance of adverse impacts associated with the project. Some environmental factors are project specific and are necessarily defined during the planning phase while others are mandated by existing regulations. forested habitat may be restricted to riparian areas and be directly and extensively impacted by project construction. et al. These physical effects tend to manifest themselves as changes in landscape diversity and associated aquatic habitat diversity or quality. dissolved oxygen. When using these systems. 6. water quality. Losses or changes in habitat will affect wildlife and aquatic life either by a reduction or change in community structure. Streambank protection projects should strive to preserve or restore existing environmental quality to the extent possible.5) to the design of riprap for the complex geometries often associated with overtopping flows. Asthetic impacts most often occur because the natural appearance of the project area is disturbed or changed and replaced by an artificial appearance due to structures or channel alignment. loss of side channels or shallow areas or replacement of natural bank with revetment. The physical impacts of streambank protection can affect channel morphology. changes in habitat composition for a specific project can be either detrimental or beneficial depending on circumstances.9. and toe of the embankment. and have been adapted to this application as a result of a recognized need. in arid regions of the western United States. whereas in the southeastern United States. 1989b) note that many of these systems were originally designed for uses other than the protection of embankments during overtopping flow. forested habitat may be more extensive.Each of these protection systems has its own unique mode of failure and threshold capability for erosion resistance.g. sides. designed and constructed with consideration being given to environmental quality factors and project objectives.. physical. In some cases streambank protection is performed in conjunction with other projects having different purposes. and biological. Water quality impacts from changes in turbidity together with alteration of riparian habitat (e. Temporary changes in water quality may occur as a result of construction activities. careful attention should be placed on the design of termination details at the crest. and it is difficult to isolate impacts due to streambank protection alone. 6. shading) affect stream temperature and photosynthetic activities that in turn may affect algae or aquatic plant populations.

becomes degraded and its recreational use is minimized. rough and irregular banks.Biological impacts can be broadly categorized into either terrestrial or aquatic. deep local scour holes. Riparian vegetation also supports aquatic species by providing habitat for these species and input to the food chain. The major terrestrial impact involves alteration or elimination of riparian zone vegetation due to construction or project features.9. The stream. The riparian zone provides and supports a wide variety of plant and animal life and often provides critical habitat for certain species. changes in land use. and economic or social impacts. with minimum local scour and no opportunity for debris jamming. (1985) concluded that most existing engineering design criteria for river training works are in direct conflict with general concepts of fish habitat maintenance since their aims are well aligned. 6. The objectives of fisheries mitigation works are often quite the opposite. the ability of the stream to cleanse itself and to assimilate wastes is lessened. may also be affected due to changes or reductions in required habitats as a result of project features. Channel stabilization can affect succession of riparian vegetation and decrease diversity. Because the functioning of the aquatic ecosystems is impaired. The chief effects of channelization are as follows: • • • • • • • Removes the natural diverse substrate materials that allow the development of many types of habitats for aquatic organisms Increases sediment load that decreases light penetration and primary production Creates a shifting bed load that is inimical to bottom-dwelling organisms Simplifies the current pattern and eliminates habitats of diverse currents Lowers the stream channel and often drains adjacent swamp areas and aquifers that help to maintain stream flow during time of low precipitation Destroys floodplain ponds that are the breeding ground for aquatic life and that act as a reservoir for species of the river proper Reduces the stability of the banks and causes cave-in of trees and other overhanging vegetation that are an important food source for stream life and whose shade reduces high stream temperatures during the summer months Kellerhals et al. instead of being one that is aesthetically pleasing and highly productive. Aquatic organisms. increased recreational pressure. Well founded guidelines for designing such diverse and irregular 6. Disturbing the system inevitably results in a reduction in diversity of species and productivity. and the improvement of water quality is slower. uniform and stable channels.54 . a degree of instability. including benthos and fish. debris jams and overhanging vegetation.2 Effects of Channelization Patrick (1973) assessed that the stream and its floodplain constitutes an integrated system that is well designed for moderating the effects of flooding waters and for maintaining high productivity in the stream. Other impacts that may occur due to streambank protection projects include loss of wetlands and historic sites.

6.32 is equivalent to Downslope flow over a plane bed inclined at an angle θ ˆ and λ = 90°.4.10 SOLVED PROBLEMS FOR STABILITY OF RIPRAP (SI) 6.Principles.F. Reference is also suggested to Rosgen (1996) and U. 1999. 2001) presents guidelines and references for bioengineering bank protection treatments. fifteen Federal agencies and partners published a manual. an oblique flow on a side slope with θ = θ and from Equation 6.9. tan φ û Using this information we wish to calculate (if the angle of repose.F. æ 1 + sin 90° + 0 ö η′ = η ç ÷=η 2 ø è It follows from Equation 6. 6. 2001) provides an introduction to stream restoration concepts and HEC-23 (Lagasse et al. in 1998. = ˆ tan φ cos θ ˆ η tan φ + sin θ Alternatively. maximum bed angle θ ˆ with η = 0. HEC-20 (Lagasse et al. For example.channels (particularly in larger rivers) are needed and can only be developed on the basis of studies with a far broader scope than the normal project-oriented work funded by developers. Army Corps of Engineers publications such as Watson et al. φ = 40°).3 that the stability factor is S.35° or 4.F. β = 0. Solving for θ 6.10.1 percent. solving for η yields ˆù é ˆ ê 1 − tan θ ú η = cos θ ë S.S. numerous agencies and practitioners have published guidelines for stream corridor restoration and channel rehabilitation design. This document represents a cooperative effort by the participating agencies to produce a common technical reference on stream corridor restoration. S. = 1 and φ = 40° yields θ ˆ = 2.1 PROBLEM 1 Stability of Particles Under Downslope Flow ˆ shown in Figure 6.55 . Stream Corridor Restoration . what is the ˆ at which η will be 5 percent different from that of a horizontal bed.3 Channel Restoration and Rehabilitation Over the last several years. Recognizing that no two stream corridors and no two restoration initiatives are identical. 6. according to Equation 6.95. Processes and Practices (Federal Interagency Stream Restoration Working Group 1998). Then. this technical document broadly addresses the elements of restoration that apply in the majority of situations encountered.

8 = = 0. Assuming angle of repose φ = 40°. The stone size is then calculated from Equation 6.65).10.36). 6.11.8 N/m .5 η= 21τ o 21 x 95.35 m (S s − 1) γη (2. at a sloping angle θ = 20°.9 are used to calculate the stability factor Sm = tan φ tan 40° = = 2.36 2 2 (b) For the same design shear stress τo = 95. Equations 6.65 − 1) 9800 x 0.04 6. 6. Definition sketch for riprap on a channel bed. and 6.8 = = 0. the stability number η is obtained from Equation 6. determine the stability factor of particle sizes Dm = 0. Dm = 21τ o 21 x 95.2 PROBLEM 2 Riprap Design on Embankment Slopes In a design situation water flows parallel to an embankment built of crushed rock riprap.10.31 x 0. (Ss = 2. calculate the riprap size that gives a stability factor equal to 1.83 sec 20 = 2.15 m.56 .3 or Figure 6.8 N/m . From Equation 6.31 tan θ tan 20° ζ = S m η sec θ = 2.15 Then. (a) If the design shear stress is τo = 95.14 for θ = 20° (η = 0. 6.Figure.83 (S s − 1) γD m (2.32.65 − 1) 9800 x 0.5.13.

66 1.30 m 6.13).24 x (2)1/ 3 = 0.65 x 9.18 and 6.3 x 3. y = 3.12 cos 20 = 0. (c) Determine riprap size Dm for a side slope.24 m D 50 = D 30 x (D 85 \ D15 )1/ 3 = 0.11. 6.31 (ζ 2 + 4) − ζ) = ((2.12.16: Sm = tan φ tan 40 = = 2.05 êç ú ÷ 0.10.1x 0.5 D 30 éæ 9800 ö 0. The following problem addresses the design of riprap for the protection of the spill through embankment.) x S m Dm = 0. S.F.5 16 .85 2. 1990). 85 x 9 .F.312 η= 2 Sm − (S.66 (d) Compare the size calculated in (c) with a riprap size calculated using the U.0 x 1.4). The side slope angle θ = 20°.S.312 − 1.57 .F. 6. very angular rock with angle of repose φ = 40°.3 V 2 0.81 x 0.1x 2.5 D 30 é Sin 2 θ ù K 1 = ê1 − 2 ú ë Sin φ û é Sin 2 20 ù = ê1 − ú 2 ë Sin 40 û 1/ 2 = 0.5.3 x 1.0 x 3. 05 ) ø è ê ú ë û = 0. Vss = 3. Army Corps of Engineers equation (Section 6.3 PROBLEM 3 Stability Factors for Riprap Design For flow around spill through abutments the angle between the horizontal and the velocity vector can be large (Figure 6.38 m (S s − 1) g η 1.5 ú è ø ë û 1/ 2 2.0.1 and D85 / D15 = 2.31 tan θ tan 20 cos θ = 2. = 1. From Equations 6. 81 x 3 . From Equations 6. and 6.66 = 1. Richardson et al.) 2 2 (S.F.66 2 = = 0.20: éæ γ ö 0 .15 m) is unstable. The draw down angle can range from 0 to 35 degrees and the reference velocity Vr in the vicinity of the riprap can be very large (Lewis 1972. 1975.66 m/s. = 1/ 2 Sm 2.04) = 0.5 ù 3.04 2 + 4)1/ 2 − 2. 5 ù V ú ÷ = S f C s C v C T y êç γs − γ ÷ êç (K 1 g y ) 0. 175 0 .94 < 1 2 2 ( ) This size fraction (Dm = 0.05 m.S. The draw down as the flow goes around the upstream end of the abutment can be very large.

This curve shows that the incipient motion rock size is approximately 0.3.F.16 η= 0.34 relates the stability factor and side slope angle of the embankment (for λ = 20°.93 η 0. determine the stability factor.203 The curve in Figure 6. this rock is more than adequate to withstand the flow velocity. By repeating the above calculations over the range of interest for Dm (with φ = 35°).4. 203 tan 35 ° î þ and from Equation 6.4° which corresponds to a 3:1 side slope. From Equation 6. The curve is obtained by employing Equations 6. If the embankment is covered with dumped rock having a specific weight Ss = 2.154 tan 35° + sin 18.0 on the 3:1 side slope.The reference velocity Vr = 1.4.4.3 is applicable and S.305 m.30 ) 1. 6. with a stability factor of 1.0305 m ft and Vr = 1.4° tan 35° = = 1.33 is obtained. and 6. ì ü ï ï cos λ ï ï −1 β = tan −1 í ý = tan 2 sin θ ï + sin λ ï ï ï η tan φ î þ ì ü ï ï cos 20 ° ï ï í ý = 11° 2 sin 18 . The velocity vector angle with the horizontal λ = 20°. the curve given in Figure 6. then Equation 6. The stability factor of a particular side slope riprap design can be increased by decreasing the side slope angle θ. 4 ° ï + sin 20° ï ï ï 0 .65 − 1) (9.16.F. If the side slope angle is decreased to zero degrees.203 í ý = 0. from Equation 6. = cos θ tan φ cos 18.6 ì1 + sin (λ + β) ü ì1 + sin (20° + 11°) ü η′ = η í ý = 0. = 1 1 = = 4.58 . and the embankment side slope angle θ = 18.203 (S s − 1) gD m (2.83 ) 2 = = 0.107 m and the maximum stability factor is less than 2.65 and an effective rock size Dm = 0. Dm = . 6.83 m/s).59 n′ tan φ + sin θ cos β 0. 6.305 ) ( ) This dumped rock has an angle of repose of approximately 35° according to Figure 3.81) (0.59.83 m/s.6 for various values of θ.30 Vr2 (0.4° cos 11° Thus. Therefore.154 2 2 î î þ þ The stability factor for the rock is given by Equation 6.3 S.

Stability factors for various rock sizes on a side slope.34. 6. Safety factors for various side slopes.33.59 . Figure 6.Figure 6.

(b) At the toe of the spill through abutment the velocity vector Vr has a magnitude of 2.16. has a velocity vector Vr with a magnitude of 1.11 6.F.3. φ ≅ 40°.16.61 tan 40 + sin 18.4.02 0. The specific gravity of the available rock is 2. and 6. slightly over 1.5 + sin 20 ï ï ï î 0.6. The riprap size is determined either by the same iterative procedure used in (a) or by using Equations 6. for the 100-year flood. S.61 2 î þ From Equation 6.091 o From Equation 6.F. 6. equals 1.16. η= 0. From Figure 3.8 ï 2 sin 18.83 2 = 0. and 6. From Equation 6.65 − 1) x 9. 6.4.8) ü η′ = 0.83 o m/s and angle with the horizontal λ = 20 .F.091 m is used. A riprap size of 0. Determine the size of riprap in this area required to resist the erosive force of water. S.6.4 PROBLEM 4 Riprap Design on an Abutment (a) Consider a spill through abutment with side slope θ = 18.68 í ý = 0.6. = cos 18.5 cos 32.5.88 m/s and an angle with the horizontal λ of 0 degrees. ì ü ï ï cos 20 ï ï β = tan −1 í ý = 32.60 . The flow near the surface. a stone size of 0.10.4.12.65.3 x 1. 6.81x .8 The procedure is repeated with increasing stone size until S.5. ì1 + sin (20 + 32. The spill slope riprap rock size is obtained by iteration from assuming Dm and calculating successively Equations 6. = 1.5 tan 40 = 1.68 tan 40 þ From Equation 6.11. Determine the size of riprap to protect the toe of the abutment.5.F.3 until the stability factor.23 m has a S.5 (3:1 side slope).68 (2. As a first approximation. Using the latter method: Equation 6.

4.4.5 = 0. Table 6. Sizes of Materials.38 m (S s − 1) g η (2.16 Dm = 0.2 η=ç ç S. 6. 5 2 ÷ è ø ö ÷ cos 18.3 x (2.38 mm D15 = 6 mm In accordance with the recommended sizes for filters: D 50 (Riprap) 12 = = 16 D 50 (Base ) 0.50 mm D85 = 24 mm D50 = 0.75 6.81x 0.25) D15 (Filter ) <5 D 85 (Base ) (6.40 ÷ ø Riprap size is obtained from Equation 6. The gradation of a filter should be such that D 50 (Filter ) < 40 D 50 (Base ) 5< D15 (Filter ) < 40 D15 (Base ) (6.7.11. Base Material (Sand) Riprap (Gravel) D85 = 1.61 .F.3 Vt2 0.5 ç 1.4 This is the size recommended.88 ) 2 = = 0.11 SOLVED PROBLEMS FOR FILTER DESIGN (SI) The requirements for a gravel filter are given in Section 6.26) 6.51 Equation 6.24) (6. S 2 m è 2 2 ö æ ÷ cos θ = ç (2.12 2 æ Sm − S.75 mm D50 = 12 mm D15 = 0. 5 x 2 .Sm = tanφ/tanθ = tan 40/tan 18. Determine if filter is needed between the riprap and the base material.5) − 1.65 − 1) x 9.1 PROBLEM 1 Filter Design The properties of the riprap and base material are given in Table 6.5 = 2.F.

17 mm D15 = 100 mm The riprap does not contain sufficient fines to act as the filter because D15 (Riprap) 100 = = 600 D15 (Base ) 0.25 D15 (Riprap) 6 = =4 D 85 (Base ) 1.25).38 which satisfies the requirement 6. the recommended upper limit (requirement 6.5) = 20 mm 6.5. Sizes of Materials.24 D15 (Riprap) 6 = = 16 D15 (Base ) 0. Determine if a filter is needed. (1968).5 mm D85 = 400 mm D50 = 0. The properties of the filter to be placed adjacent to the base are as follows: (1) so D 50 (Filter ) < 40 D 50 (Base ) D50 (Filter) < (40) (0.5 mm D50 = 200 mm D15 = 0.62 .26).11.5.17 which is much greater than 40.5 which is much greater than 5.26.which satisfies requirement 6. the recommended upper limit (requirement 6.2 PROBLEM 2 Filter Design The following filter design is taken from Anderson et al. Table 6.5 which satisfies the requirement 6. Also D15 (Riprap) 100 = = 67 D 85 (Base ) 1. Base Material (Sand) Riprap (Rock) D85 = 1. 6. The riprap itself satisfies the requirements for the filter so no filter is needed. The properties of the base material and the riprap are given in Table 6.

85 mm Thus.(2) so (3) so (4) so D15 (Filter ) < 40 D15 (Base ) D15 (Filter) < (40) (0.5 mm D15 (Filter ) >5 D15 (Base ) D15 (Filter) > (5) (0.17) = 6.17) = 0.63 .5) = 7.8 mm D15 (Filter ) <5 D 85 (Base ) D15 (Filter) < (5) (1.8 mm and D50 (Filter) < 2 mm The properties of the filter to be placed adjacent to the riprap are as follows: (1) D 50 (Riprap) < 40 D 50 (Filter ) D 50 (Filter ) > 200 = 5 mm 40 so (2) so D15 (Riprap) >5 D15 (Filter ) 100 D15 (Filter ) > = 20 mm 5 D15 (Riprap) < 40 D15 (Filter ) D15 (Filter ) > 100 = 2.85 mm < D15 (Filter) < 6. with respect to the base 0.5 mm 40 (3) so (4) D15 (Riprap) <5 D 85 (Filter ) 6.

12. a good filter could have these sizes: D85 = 40 mm D50 = 10 mm D15 = 4 mm Figure 6.so D 85 (Filter ) > 100 = 20 mm 5 Therefore. an oblique flow on a side slope with θ = θ and from Equation 6. with respect to the riprap. Then. Gradations of filter blanket for Problem 2 (after Anderson et al. the filter must satisfy these requirements 2.12 SOLVED PROBLEMS FOR STABILITY OF RIPRAP (ENGLISH) 6. β = 0.4.35. Any filter having sizes represented by the double cross-hatched area is satisfactory.35. For example. according to Equation 6.6.64 . 6.5mm < D15 (Filter) < 20 mm D50 (Filter) > 5 mm D85 (Filter) > 20 mm These riprap filter requirements along with those for the base material are shown in Figure 6.1 PROBLEM 1 Stability of Particles Under Downslope Flow ˆ shown in Figure 6.36 is equivalent to Downslope flow over a plane bed inclined at an angle θ ˆ and λ = 90°. 6. 1968).

F.95.12. at a sloping angle θ = 20°. (a) If the design shear stress is τo = 2. Solving for θ 6.æ 1 + sin 90° + 0 ö η′ = η ç ÷=η 2 ø è It follows from Equation 6. S.F. 6. solving for η yields ˆù é ˆ ê 1 − tan θ ú η = cos θ ë S.5. φ = 40°).35° or 4.65).F.1 percent. Using this information we wish to calculate (if the angle of repose.0 psf.3 that the stability factor is S. Definition sketch for riprap on a channel bed. 6.65 . tan φ û Figure. maximum bed angle θ ˆ with η = 0. (Ss = 2. calculate the riprap size that gives a stability factor equal to 1. = 1 and φ = 40° yields θ ˆ = 2.36.2 PROBLEM 2 Riprap Design on Embankment Slopes In a design situation water flows parallel to an embankment built of crushed rock riprap. = ˆ tan φ cos θ ˆ η tan φ + sin θ Alternatively. what is the ˆ at which η will be 5 percent different from that of a horizontal bed.

5.02 2 + 4)1/ 2 − 2. 6.4).Assuming angle of repose φ = 40°.02 S.65 − 1) 62. the stability number η is obtained from Equation 6. = 1. The side slope angle θ = 20°.3 V 2 0.) x S m Dm = 0 .82 sec 20 = 2. 6.16 Sm = tan φ tan 40 = = 2.F.3 or Figure 6.1 and D85 / D15 = 2.5 ú è ø ë û 2. 5 ù V ú ÷ = S f C s C v C T y êç γs − γ ÷ êç (K 1 g y ) 0.13 ft (S s − 1) γη (2.5 D 30 6.4 x 0.2 x 0.12 cos 20 = 0.25 ft (S s − 1) g η 1.12. Army Corps of Engineers equation (Section 6.31 (ζ 2 + 4) − ζ) = ((2.3 x 12 2 = = 1.65 − 1) 62.66 . very angular rock with angle of repose φ = 40°.F.31 tan θ tan 20° ζ = S m η sec θ = 2.66 (d) Compare the size calculated in (c) with a riprap size calculated using the U.5 Then.31 tan θ tan 20 cos θ = 2.95 < 1 2 2 ( ) This size fraction (Dm = 6 inches) is unstable.20: éæ γ ö 0 .82 (S s − 1) γD m (2. S.312 = 1. Dm = 21τ o 21 x 2.4 x 0.13. and 6.F.36).) 2 Sm 2 (S. determine the stability factor of particle sizes Dm = 6 inches.18 and 6.F. = 1/ 2 Sm 2. y = 10 ft.66 1.5 η= 21τ o 21 x 2 = = 0. From Equations 6. From Equation 6.36 (b) For the same design shear stress τo = 2.65 x 32.312 η= 2 − (S.S.11.0 = = 1.0.31 x 0. and 6.1x 2. From Equations 6.11.02) = 0.10.0 psf. Equations 6.9 are used to calculate the stability factor Sm = tan φ tan 40° = = 2. (c) Determine riprap size Dm for a side slope. The stone size is then calculated from Equation 6.14 for θ = 20° (η = 0. Vss = 12 fps.

79 ft D 50 = D 30 x (D 85 / D15 )1/ 3 = 0.0 ft.3 6. Therefore.00 ft 6. If the embankment is covered with dumped rock having a specific weight Ss = 2. Richardson et al. determine the stability factor. The draw down as the flow goes around the upstream end of the abutment can be very large.6 ì1 + sin (λ + β) ü ì1 + sin (20° + 11°) ü η′ = η í ý = 0.4° + sin 20° ï ï 0.0 x 10 êç ú ÷ 0.é Sin 2 θ ù = K 1 ê1 − 2 ú ë Sin φ û 1/ 2 é Sin 2 20 ù = ê1 − ú 2 ë Sin 40 û 1/ 2 = 0. and the embankment side slope angle θ = 18. from Equation 6.0 x 1.65 − 1) (32.16 0.30 ) (6) 2 η= = = 0.3 PROBLEM 3 Stability Factors for Riprap Design For flow around spill through abutments the angle between the horizontal and the velocity vector can be large (Figure 6.5 ù 12 = 1.154 2 2 î î þ þ The stability factor for the rock is given by Equation 6.5 ê ëè 102.12.203 í ý = 0.2 x 10 ) ú û = 0.30 Vr2 (0.67 . 1975. The velocity vector angle with the horizontal λ = 20°.4.203 (S s − 1) gD m (2. 5 D 30 éæ 62. From Equation 6. 1990).85 x 32.4.4° which corresponds to a 3:1 side slope.1x 0.65 and an effective rock size Dm = 1. The draw down angle can range from 0 to 35 degrees and the reference velocity Vr in the vicinity of the riprap can be very large (Lewis 1972.85 2.4 ö 0.0 ) ( ) This dumped rock has an angle of repose of approximately 35° according to Figure 3. The following problem addresses the design of riprap for the protection of the spill through embankment.2) (1.79 x (2)1/ 3 = 1.96 ø 0. ì ü ï ï cos λ ï ï −1 β = tan −1 í ý = tan 2 sin θ ï + sin λ ï ï η tan φ ï î þ ì ü ï ï cos 20° ï ï í ý = 11° ï 2 sin 18.3 x 1.203 tan 35° ï î þ and from Equation 6. The reference velocity Vr = 6 fps.13).

S. Determine the size of riprap in this area required to resist the erosive force of water.203 The curve in Figure 6. = 1 1 = = 4.3 is applicable and S. and 6. with a stability factor of 1. then Equation 6.F.37.38 relates the stability factor and side slope angle of the embankment (for λ = 20°. Figure 6. If the side slope angle is decreased to zero degrees.4 PROBLEM 4 Riprap Design on an Abutment (a) Consider a spill through abutment with side slope θ = 18.107 m and the maximum stability factor is less than 2. 6. The flow near the surface.65. Dm = 1.59.0 ft and Vr = 6. The stability factor of a particular side slope riprap design can be increased by decreasing the side slope angle θ.4° cos 11° Thus. this rock is more than adequate to withstand the flow velocity.68 .4.4° tan 35° = = 1.F. 6.6 for various values of θ. = cos θ tan φ cos 18.3.5 (3:1 side slope).59 n′ tan φ + sin θ cos β 0.0 fps). the curve given in Figure 6. Stability factors for various rock sizes on a side slope. has a velocity vector Vr with a magnitude of 6.12.37 is obtained. 6.0 on the 3:1 side slope. This curve shows that the incipient motion rock size is approximately 0. The curve is obtained by employing Equations 6.16.0 o fps and angle with the horizontal λ = 20 . The specific gravity of the available rock is 2. for the 100-year flood.93 η 0. By repeating the above calculations over the range of interest for Dm (with φ = 35°). o 6.154 tan 35° + sin 18.

6.3 ft is used.3 until the stability factor. and 6.69 .68 tan 40 þ From Equation 6.5 + sin 20 ï ï ï î 0.6. S.8 ï 2 sin 18.2) x (0.4. φ ≅ 40°.3) Figure 6.F. η= 0 . From Figure 3. equals 1.The spill slope riprap rock size is obtained by iteration from assuming Dm and calculating successively Equations 6. 6.5. As a first approximation. ì ü ï ï cos 20 ï ï β = tan −1 í ý = 32. 6. a stone size of 0.38.16.6. Safety factors for various side slopes.68 (2.4.3 x 6 2 = 0.4.16. From Equation 6.65 − 1) x (32. From Equation 6.

S. Using the latter method: Equation 6.45 fps and an angle with the horizontal λ of 0 degrees.26 ft (S s − 1) g η (2.5.8 The procedure is repeated with increasing stone size until S.12.3 x (9.F. A riprap size of 9 inches would give a S. The riprap size is determined either by the same iterative procedure used in a or by using Equations 6. slightly over 1.02 0.12 2 æ Sm − S.61 tan 40 + sin 18.ì1 + sin (20 + 32.F. and 6. 6.F. 5 2 ÷ è ø ö ÷ cos 18.16.5 ç 1.11.65 − 1) x 32.5 = 0.4 This is the size recommended.7.3 Vr2 0.68 í ý = 0.40 ÷ ø Riprap size is obtained from Equation 6.13 SOLVED PROBLEMS FOR FILTER DESIGN (ENGLISH) The requirements for a gravel filter are given in Section 6. (b) At the toe of the spill through abutment the velocity vector Vr has a magnitude of 9.2 η=ç ç S.70 . S 2 m è 2 2 ö æ ÷ cos θ = ç (2.51 Equation 6.8) ü η′ = 0. Determine the size of riprap to protect the toe of the abutment.5 cos 32. = cos 18. The gradation of a filter should be such that D 50 (Filter ) < 40 D 50 (Base ) (6.11 Sm = tanφ/tanθ = tan 40/tan 18.5) − 1.F.2 x 0. = 1.45) 2 = = 1.16 Dm = 0. 6.5 tan 40 = 1.F.61 2 î þ From Equation 6.24) 6.5. 5 x 2 .5 = 2.3.

5< D15 (Filter ) < 40 D15 (Base ) (6. Determine if a filter is needed.25 D15 (Riprap) 6 = =4 D 85 (Base ) 1.38 which satisfies the requirement 6. 6. The riprap itself satisfies the requirements for the filter so no filter is needed.13. sediment sizes are given in mm for the English system. 6.25) D15 (Filter ) <5 D 85 (Base ) (6.24 D15 (Riprap) 6 = = 16 D15 (Base ) 0.75 mm D50 = 12 mm D15 = 0. The properties of the base material and the riprap are given in Table 6.26) 6. Base Material (Sand) Riprap (Gravel) D85 = 1.26. (1968).5 which satisfies the requirement 6. Sizes of Materials.75 which satisfies requirement 6.6.2 PROBLEM 2 Filter Design The following filter design is taken from Anderson et al. Determine if filter is needed between the riprap and the base material.38 mm D15 = 6 mm In accordance with the recommended sizes for filters: D 50 (Riprap) 12 = = 16 D 50 (Base ) 0.6.13. Note: Generally. Table 6.71 .50 mm D85 = 24 mm D50 = 0.7.1 PROBLEM 1 Filter Design The properties of the riprap and base material are given in Table 6.

8 mm D15 (Filter ) <5 D 85 (Base ) D15 (Filter) < (5) (1.85 mm < D15 (Filter) < 6.5 mm D50 = 200 mm D15 = 0.72 .5 mm D85 = 400 mm D50 = 0.17) = 0.85 mm Thus.17) = 6. Base Material (Sand) Riprap (Rock) D85 = 1. the recommended upper limit (requirement 6.5) = 20 mm D15 (Filter ) < 40 D15 (Base ) D15 (Filter) < (40) (0.8 mm 6.7.26). Sizes of Materials. The properties of the filter to be placed adjacent to the base are as follows: (1) so (2) so (3) so (4) so D 50 (Filter ) < 40 D 50 (Base ) D50 (Filter) < (40) (0.17 mm D15 = 100 mm The riprap does not contain sufficient fines to act as the filter because D15 (Riprap) 100 = = 600 D15 (Base ) 0.5 mm D15 (Filter ) >5 D15 (Base ) D15 (Filter) > (5) (.5) = 7.Table 6. with respect to the base 0. Also D15 (Riprap) 100 = = 67 D 85 (Base ) 1.5 which is much greater than 5.17 which is much greater than 40.25). the recommended upper limit (requirement 6.

For example.73 .5mm < D15 (Filter) < 20 mm D50 (Filter) > 5 mm D85 (Filter) > 20 mm These riprap filter requirements along with those for the base material are shown in Figure 6.39. with respect to the riprap. Any filter having sizes represented by the double cross-hatched area is satisfactory. a good filter could have these sizes: D85 = 40 mm D50 = 10 mm D15 = 4 mm 6.5 mm 40 (3) so (4) D15 (Riprap) <5 D 85 (Filter ) D 85 (Filter ) > 100 = 20 mm 5 so Therefore.and D50 (Filter) < 2 mm The properties of the filter to be placed adjacent to the riprap are as follows: (1) D 50 (Riprap) < 40 D 50 (Filter ) D 50 (Filter ) > 200 = 5 mm 40 so (2) so D15 (Riprap) >5 D15 (Filter ) 100 D15 (Filter ) > = 20 mm 5 D15 (Riprap) < 40 D15 (Filter ) D15 (Filter ) > 100 = 2. the filter must satisfy these requirements 2.

Gradations of filter blanket for Problem 2 (after Anderson et al. 6.Figure 6. 1968).74 .39.

CHAPTER 7 SCOUR AT BRIDGES

7.1 INTRODUCTION

7.1.1 General Scour at highway structures is the result of the erosive action of flowing water removing bed material from around the abutments and piers which support the bridge and bed and bank material of the stream the structure crosses. Both scour at highway structures and stream migration (instability) can cause a bridge failure. All material in a streambed will erode. It is just a matter of time. However, some material such as granite may take hundred's of years to erode. Whereas, sandbed streams will erode to the maximum depth of scour in hours. Sandstone, shales, and other sedimentary bedrock materials, although they will not erode in hours or even days will, over time, if subjected to the erosive forces of water, erode to the extent that a bridge will be in danger unless the substructures are founded deep enough. Cohesive bed and bank material such as clays, silty clays, silts and silty sands or even coarser bed material such as glacial tills, which are cemented by chemical action or compression, will erode if subjected to the forces of flowing water. The erosion of cohesive and other cemented material is slower than sand bed material, but their ultimate scour will be as deep if not deeper than the scour depth in a non-cohesive sandbed stream (Briaud et al. 1999). It might take the erosive action of several major floods but ultimately the scour hole will be equal to or greater in depth than with a sand bed material. This does not mean that every bridge foundation must be buried below the calculated scour depth determined for non-bedrock streams. It does mean that so-called bedrock streams must be carefully evaluated. Scour at bridge crossings is a sediment transport process. Long-term degradation, general scour, and local scour at piers and abutments result from the fact that more sediment is removed from these areas than is transported into them. If there is no transport of bed material into the bridge crossing, clear-water scour exists. Transport of appreciable bed material into the crossing results in live-bed scour. In this latter case the transport of the bed material limits the scour depth. Whereas, with clear-water scour the scour depths are limited by the critical velocity or critical shear stress of a dominant size in the bed material at the crossing. 7.1.2 Costs of Bridge Failure from Scour Hydraulic factors (scour/ice/debris) cause 60 percent of bridge failures in the United States (Shirole and Holt 1991). In the United States there are over 580,000 bridges in the National Bridge Inventory. These numbers include federal highway system, state, county and city bridges. Approximately 84 percent of these bridges are over water. Bridge failures cost millions of dollars each year as a result of both direct cost necessary to replace and restore bridges, and indirect costs related to disruption of transportation facilities. However, of even greater consequence is loss of life from bridge failures. Chang, in a 1973 study for the

7.1

Federal Highway Administration, indicated that about $75 million was expended annually up to 1973 to repair roads and bridges that were damaged by floods. This cost does not include the additional indirect costs to highway users for fuel and operating costs resulting from temporary closure and detours and to the public for costs associated with higher tariffs, freight rates, additional labor costs and time. Rhodes and Trent (1999) document that $1.2 billion was expended for restoration of flood damaged highway facilities during the 1980s. They state that this amount is conservative because it (1) only includes the amount funded by the U.S. Government which ranges from 75 to 100 percent of the total restoration costs, and (2) the funds were only for disasters related to very large floods and do not include the hundreds of smaller events that occur every year. They also demonstrate that the added cost of operating a vehicle over a detour and time lost traveling when a bridge failed (only part of the indirect costs) exceed by several times the direct cost of bridge replacement or repair.

7.1.3 Bridge Scour Evaluation Program The U.S. Department of Transportation, Federal Highway Administration (FHWA) in 1988 issued Technical Advisory T5140.20 (superseded by T5140.23 in 1991) requiring the States to conduct a scour evaluation program. This program resulted from the failure of the I-90 bridge over Schoharie Creek in upstate New York which killed ten people (NTSB 1988 and Richardson et al. 1987). The evaluation is to be conducted by an interdisciplinary team of hydraulic, geotechnical and structural engineers who can make the necessary engineering judgments to determine the vulnerability of a bridge to scour. As of February 2001, the 481,313 bridges over water in the National Bridge Inventory have been screened as to their scour vulnerability (100 percent). The evaluation program in the U.S. is on schedule and scour countermeasures are being implemented on bridges that have been identified as scour susceptible or scour critical. Replacement bridges are being constructed as rapidly as funds can be provided. Scour countermeasures include riprap protection, scour monitoring before, during and after a flood and the inspection program (Richardson and Davis 2001, Lagasse et al. 1997, 2001a, and 2001b, Schall et al. 1997a and 1997b).

7.1.4 Comprehensive Scour Analysis This chapter presents background knowledge on scour analysis. Earlier chapters give a comprehensive overview of fluvial geomorphology, sediment transport, and flow in alluvial channels and stream stability fundamentals needed to understand streams, stream instability and scour. The comprehensive procedure for scour and stream instability analyses and countermeasures to control them are given in the FHWA publications HEC-18, HEC-20, and HEC-23 (Richardson and Davis 2001; Lagasse et al. 2001a, 2001b). The interrelationships and procedures recommended in the three documents are shown in the Figure 7.1.

7.2

**HEC-20 Stream Stability and Geomorphic Assessment
**

Office Data Collection & Site Visit

**HEC-18 Hydrologic, Hydraulic and Scour Analysis
**

Hydrologic Analysis

**HEC-23 Bridge Scour and Stream Instability Countermeasures
**

Develop Plan of Action Evaluate CM Options with Matrix

Figure 7.1. Flow chart for scour and stream stability analysis and countermeasures.

Yes

Sufficient Data No Field Data Collection

Hydraulic Analysis

Riverine Structural / Geotechnical Input

Tidal

Hydraulic

Select CM Type

Structural

Scour Analysis

Monitoring Design CM / Monitoring Plan

Define / Classify Stream

7.3

Plot Scour Prism Evaluate Stream Stability Multi Disciplinary Evaluation Hydraulics/Structures/Geotechnical

Assess Stream Response Establish Level of Analysis

Environmental Considerations / Permitting

Evaluate CM Impact Low Risk Yes Structure Stable No No (New Bridge, revise design)

Bridge Type Existing Low Risk Scour No Susceptible

New

(Existing Bridge)

Acceptable CM Impact Yes Install CM / Implement Monitoring Plan

No

Scour Critical: Plan of Action Required

Yes

Replace Bridge

No

CM Viable

Yes Inspection & Maintenance

7.2 TOTAL SCOUR Total scour at a highway crossing is composed of three components: (1) Long-Term Aggradation or Degradation The change in river bed elevation (aggradation or degradation) over long lengths and time due to changes in controls, such as dams, changes in sediment discharge, head cuts, daily tidal flow, and changes in river geomorphology, such as changing from a meandering to a braided stream. These processes may be natural or human induced. (2) General Scour The scour that results from the acceleration of the flow due to either a natural or bridge contraction or both (contraction scour). General scour may also result from the location of the bridge on the stream. For example, its location with respect to a stream bend or its location upstream from the confluence with another stream. In this latter case, the elevation of the downstream water surface will affect the backwater on the bridge, hence, the velocity and scour. General scour may occur during the passage of a flood and the stream may fill in on the falling stage. This type of scour involves the removal of material from the bed and banks across all or most of the width of a channel. (3) Local Scour The scour that occurs at a pier or abutment as the result of the pier or abutment obstructing the flow. These obstructions accelerate the flow and create vortices that remove the material around them. Generally, scour depths from local scour are much larger than long-term degradation or general scour, often by a factor of ten. But, if there are major changes in the stream conditions, such as a large dam built upstream or downstream of the bridge or severe straightening of the stream, long-term bed elevation changes can be the larger element in the total scour. Also, scour depths from severe contraction of the flow, (often causing ponding upstream of the bridge) can be larger than local scour. (4) Lateral Shifting of the Stream In addition to the above, lateral shifting of the stream may also erode the approach roadway to the bridge and, by changing the angle of the flow in the waterway at the bridge crossing, change the total scour. 7.3 CLEAR-WATER AND LIVE-BED SCOUR The two conditions of general and local scour are (1) clear-water scour and (2) live-bed scour. Clear-water scour occurs when there is no movement of the bed material in the main channel of the stream upstream of the crossing, or the sediment transport in the upstream reach or floodplain is transported through the bridge opening or local scour holes in suspension. The increase in velocity by contraction of the flow by the bridge or the acceleration of the flow and vortices created by the piers or abutments causes the bed material in the bridge opening or at their base to move.

7.4

Live-bed scour occurs when the bed material upstream of the crossing is moving and moves as contact sediment discharge in the contracted bridge opening and/or into the local scour holes. Typical clear-water scour situations include (1) coarse bed material streams, (2) flat gradient streams during low flow, (3) local deposits of larger bed materials that are larger than the biggest fraction being transported by the flow (rock riprap is a special case of this situation), (4) armored stream beds where the only locations that tractive forces are adequate to penetrate the armor layer are at piers and/or abutments, (5) vegetated channels where, again, the only locations that the cover is penetrated is at piers and/or abutments, and (6) streams with fine bed material and large velocity or shear stress whereby the bed material in transport washes through a contraction or local scour hole Clear-water scour reaches its maximum over a longer period of time than live-bed scour (Figure 7.2). This is because clear-water scour occurs mainly on coarse bed material streams. In fact, clear-water scour may not reach its maximum until after several floods. Also, maximum clear-water scour is about 10 percent greater than the equilibrium live-bed scour. Bridges over coarse bed material streams often have clear-water scour at the lower part of a hydrograph, live-bed scour at the higher discharges, and then clear-water scour on the falling stages.

Figure 7.2. Local scour depth at a pier as a function of time.

Live-bed scour in sand bed streams with a dune bed configuration fluctuates about an equilibrium scour depth (Figure 7.2). The reason for this is the fluctuating nature of the sediment transport of the bed material in the approaching flow when the bed configuration of the stream is dunes. In this case (dune bed configuration in the channel upstream of the bridge), maximum depth of scour is about 30 percent larger than equilibrium depth of scour.

7.5

The maximum depth of scour is the same as the equilibrium depth of scour for live-bed scour with a plane bed configuration. With antidunes occurring upstream and in the bridge crossing; the maximum depth of scour, from the limited research of Jain and Fisher (1979), is from 10 to 20 percent greater than the equilibrium depth of scour. In general, with sand bed streams a dune bed changes to plane bed or antidune flow during flood flow. 7.4 ARMORING Armoring occurs on a stream or in a scour hole when the forces of the water during a particular flood are unable to move the larger sizes of the bed material. This protects the underlying material from movement. Contraction scour or local scour around an abutment or pier may initially occur, but as the scour depth increases the coarser bed material moves down in the bridge cross-section or local scour hole and protects the bed so that the full scour potential is not reached. When armoring occurs, the coarser bed material will tend to remain in place or quickly redeposit so as to form a layer of riprap-like armor in the cross-section or local scour hole and, thus, limit further scour for a particular discharge. When larger flows occur the armor layer can be broken and the scour depth deepened either until a new armor layer is developed or the maximum scour is reached. 7.5 LONG-TERM BED ELEVATION CHANGES Long-term bed elevation changes (aggradation or degradation) may be the natural trend of the stream or may be the result of some modification to the stream or watershed condition. In scour analyses, only long-term degradation is considered. Aggradation, if it occurs, is not used to decrease total scour estimate. The streambed may be aggrading, degrading or not changing (equilibrium) in the bridge crossing reach. When the bed of the stream is neither aggrading or degrading, it is in equilibrium with the sediment discharge supplied to the bridge reach and the elevation of the bed does not change. In this section, we consider long-term trends, not the cutting and filling of the bed of the stream that might occur during a runoff event. A stream may cut and fill during a runoff event and also have a long-term trend of an increase or decrease in bed elevation. The problem for the engineer is to determine what the long-term bed elevation changes will be during the lifetime of the structure. What is the current rate of change in the stream bed elevation? Is the stream bed elevation in equilibrium? Is the streambed degrading? Is it aggrading? Is there a head cut or nickpoint moving upstream? What is the future trend in the stream bed elevation? During the life of the bridge the present trend may change. These long-term changes are the result of modifications of the state of the stream or watershed. Such changes may be the result of natural processes of the result of human activities. The engineer must assess the present state of the stream and watershed and determine future changes in the river system, and from this assessment determine the long-term stream bed elevation. Factors that affect long-term bed elevation changes are: dams and reservoirs (up or downstream of the bridge), changes in watershed land use (urbanization, deforestation, etc.), channelization, cutoff of a meander bend (natural or human induced), changes in the

7.6

downstream base level (control) of the bridge reach, gravel mining from the stream bed, diversion of water into or out of the stream, natural lowering of the total system, movement of a bend, bridge location in reference to stream planform, and stream movement in relation to the crossing. Analysis of long-term streambed elevation changes must be made using the principles of river mechanics in the context of a fluvial system analysis. Such analysis of a fluvial system requires the consideration of all influences upon the bridge crossing, i.e., runoff from the watershed to the channel (hydrology), the sediment delivery to the channel (erosion), the sediment transport capacity of the channel (hydraulics) and the response of the channel to these factors (geomorphology and river mechanics). Many of the largest impacts are from human activities, either in the past, the present or the future. The analysis requires a study of the past history of the river and human activities on it; a study of present water and land use and stream control activities; and, finally, contacting all agencies involved with the river to determine future changes in the river. A method to organize such an analysis is to use a three level fluvial system approach (see Chapter 9 and HEC-20). This method provides three levels of detail in an analysis: (1) a qualitative determination based on general geomorphic and river mechanics relationships; (2) engineering geomorphic analysis using established qualitative and quantitative relationships to establish the probable behavior of the stream system to various scenarios of future conditions; and (3) quantifying the changes in bed elevation using available physical process mathematical models such as BRI-STARS (Molinas 2000) or HEC-6, (U.S. Army Corps of Engineers 1993). Methods to be used in the three levels of analysis are given in this manual, FHWA’s HEC-18 (Richardson and Davis 2001) and HEC-20 (Lagasse et al. 2001). 7.6 GENERAL SCOUR General scour at a bridge can be caused by a decrease in channel width, either naturally or by the bridge, which decreases flow area and increases velocity. This is contraction scour. General scour can also be caused by short-term (daily, weekly, yearly or seasonally) changes in the downstream water surface elevation that controls the backwater and hence the velocity through the bridge opening. Because this scour is reversible it is included in general scour rather than in long-term scour. General scour can result from the location of the bridge with regard to a bend. If the bridge is located on or close to a bend, the concentration of the flow on the outer part of the channel can erode the bed. General scour can be cyclic. That is, during a runoff event the bed scours during the rise in stage (increasing discharge) and fills on the falling stage (deposition). General scour from a contraction occurs when the flow area of a stream is decreased from the normal either by a natural constriction or by a bridge. With the decrease in flow area there is an increase in average velocity and bed shear stress. Hence, there is an increase in stream power at the contraction and more bed material is transported through the contracted reach than is transported into the reach. The increase in transport of bed material lowers the bed elevation. As the bed elevation is lowered, the flow area increases and the velocity and shear stress decreases until equilibrium is reached, that is the bed material transported into the reach is equal to that which is transported out of the reach.

7.7

The contraction of the flow by the bridge can be caused by a decrease in flow area of the stream channel by the abutments projecting into the channel and/or the piers taking up a large portion of the flow area. Also, the contraction can be caused by the approaches to the bridge cutting off the overland flow that normally goes across the floodplain during high flow. This latter case causes clear-water scour at the bridge section because the overland flow normally does not transport any bed material sediments. This clear-water flow picks up additional sediment from the bed when it returns to the bridge crossing. In addition, if it returns to the stream channel at an abutment can increase the local scour there. A guide bank at that abutment decreases the risk from scour from this returning overbank flow. Also, relief bridges in the approaches decrease the scour problem at the bridge cross section by decreasing the amount of flow returning to the natural channel. Other Factors that can cause contraction scour are: (1) (2) (3) (4) (5) (6) (7) a natural stream constriction long approaches over the floodplain to the bridge ice formation or jams berm forming along the banks by sediment deposits island or bar formations upstream or downstream of the bridge opening debris growth of vegetation in the channel or floodplain

To determine the magnitude of general scour from a variable backwater requires a study of the stream system to (1) determine if this condition exists and (2) determine the magnitude of general scour for this condition. Of particular value in determining if backwater effects exist and the magnitude of the effects on the velocity and depth is the WSPRO computer model. The difference in depth between the highest expected bed elevation and the lowest expected bed elevation for the design discharge is the value of the general scour. General scour of the bridge opening may be concentrated in one area. If the bridge is located on or close to a bend the scour will be concentrated on the outer part of the bend. In fact, there may be deposition on the inner portion of the bend, further concentrating the flow, which increases the scour at the outer part of the bend. Also at bends, the thalweg (the part of the stream where the flow or velocity is largest) will shift toward the center of the stream as the flow increases. This can increase scour and the non-uniform distribution of the scour in the bridge opening. Often the magnitude of general scour cannot be predicted and inspection is the solution for general scour problems. Also, a physical model study can be used to determine general scour.

7.6.1 Contraction Scour Contraction Scour Conditions. Contraction scour equations are based on the principle of conservation of sediment transport (continuity). In the case of live-bed scour, the fully developed scour in the bridge cross section reaches equilibrium when sediment transported into the contracted section equals sediment transported out. As scour develops, the shear stress in the contracted section decreases as a result of a larger flow area and decreasing average velocity. For live-bed scour, maximum scour occurs when the shear stress reaches the point that bed-material transported in equals the bed-material transported out and the conditions for sediment continuity are in balance. For clear-water scour, the bed-material

7.8

transported into the contracted section is essentially zero and maximum scour occurs when the shear stress or velocity reaches the critical shear stress or critical velocity of the bed material in the section. Chapter 3 and HEC-18 give equations and methods for calculating critical shear stress or critical velocity. Live-bed contraction scour occurs at a bridge when there is transport of bed material in the upstream reach into the bridge cross section. With live-bed contraction scour, the area of the contracted section increases until, in the limit, the transport of bed material out of the contracted section equals the bed material transported in. Normally, the width of the contracted section is constrained and depth increases until the limiting conditions are reached. Clear-water contraction scour occurs in a long contraction when (1) there is no bed material transport from the upstream reach into the downstream reach or (2) the material being transported in the upstream reach is transported through the downstream reach mostly in suspension and at less than capacity of the flow. With clear-water contraction scour, the area of the contracted section increases until, in the limit, the velocity (V) of the flow or the shear stress (τo) on the bed is equal to the critical velocity (Vc) or the critical shear stress (τc) of a certain particle size (D) in the bed material. Normally, the width (W) of the contracted section is constrained and the depth (y) increases until the limiting conditions are reached. Live-bed scour depths may be limited if there are appreciable amounts of large-sized particles in the bed material. It is appropriate, then, to use the clear-water scour equation in addition to the live-bed scour equation and use the smaller of the two depths. Also, it is appropriate to use the clear-water scour equation if the transport of bed material from upstream of the contraction is small in quantity or composed of fine material that washes through the contraction in suspension. There are four conditions (cases) of contraction scour at bridge sites depending on the type of contraction, and whether there is overbank flow or relief bridges. Regardless of the case, contraction scour can be evaluated using two basic equations: (1) live-bed scour, and (2) clear-water scour. For any case or condition, it is only necessary to determine if the flow in the main channel or overbank area upstream of the bridge, or approaching a relief bridge, is transporting bed material (live-bed) or is not (clear-water), and then apply the appropriate equation with the variables defined according to the location of contraction scour (channel or overbank). Critical Velocity. To determine if the flow upstream of the bridge is transporting bed material, calculate the critical velocity for beginning of motion Vc of the D50 size of the bed material and compare it with the mean velocity V of the flow in the main channel or overbank area upstream of the bridge opening. If the critical velocity of the bed material is larger than the mean velocity (Vc > V), then clear-water contraction scour will exist. If the critical velocity is less than the mean velocity (Vc < V), then live-bed contraction scour will exist. To calculate the critical velocity use the equation derived in Chapter 3. This equation is:

Vc = K u y 1/ 6 D1/ 3

(7.1)

where: Vc y = = Critical velocity above which bed material of size D and smaller will be transported, m/s (ft/s) Depth of flow, m (ft)

7.9

D D50 Ku Ku Ku

= = = = =

Particle size for Vc, m (ft) Particle size in a mixture of which 50 percent are smaller, m (ft) Coefficient derived in Chapter 3 6.19 SI units 11.25 English units

Live-Bed Contraction Scour. A modified version of Laursen's 1960 equation for live-bed scour at a long contraction is recommended to predict the depth of scour in a contracted section. The modification is to eliminate the ratio of Manning's n (see the following Note #3).

y 2 éQ2 ù =ê ú y 1 ë Q1 û

6/7

é W1 ù ê ú ë W2 û

k1

(7.2)

y s = y 2 − y o = average scour depth

(7.3)

where: y1 y2 yo Q1 Q2 W1 W2 k1 = = = = = = = = Average depth in the upstream main channel, m (ft) Average depth in the contracted section, m (ft) Existing depth in the contracted section before scour, m (ft) (see Note 7) Flow in the upstream channel transporting sediment, m3/s (ft3/s) Flow in the contracted channel, m3/s (ft3/s) Bottom width of the upstream main channel, m (ft) Bottom width of the main channel in the contracted section less pier width(s), m (ft) Exponent determined below k1 0.59 0.64 0.69 Mode of Bed Material Transport Mostly contact bed material discharge Some suspended bed material discharge Mostly suspended bed material discharge

V*/ω <0.50 0.50 to 2.0 >2.0 V* ω g S1 τo ρ Notes: = = = = = =

(τo/ρ)½ = (gy1 S1)½, shear velocity in the upstream section, m/s (ft/s) Fall velocity of bed material based on the D50, m/s (Figure 3.1) For fall velocity in English units (ft/s) multiply fall velocity in m/s by 3.28 Acceleration of gravity (9.81 m/s2) (32.2 ft/s2) Slope of energy grade line of main channel, m/m (ft/ft) Shear stress on the bed, Pa (N/m2) (lb/ft2) Density of water (1000 kg/m3) (1.94 slugs/ft3)

1. Q2 is the total flow going through the bridge opening. 2. Q1 is the flow in the main channel upstream of the bridge, not including overbank flows. 3. The Manning’s n ratio is eliminated in Laursen's live-bed equation to obtain Equation 7.2. This was done for the following reasons. The ratio can be significant for a condition of

7.10

dune bed in the main channel and a corresponding plane bed, washed out dunes or antidunes in the contracted channel. However, Laursen's equation does not correctly account for the increase in transport that will occur as the result of the bed planing out (which decreases resistance to flow, increases the velocity and the transport of bed material at the bridge). That is, Laursen's equation indicates a decrease in scour for this case, whereas in reality, there would be an increase in scour depth. In addition, at flood flows, a plane bedform will usually exist upstream and through the bridge waterway, and the values of Manning’s n will be equal. 4. W1 and W 2 are not always easily defined. In some cases, it is acceptable to use the topwidth of the main channel to define these widths. Whether topwidth or bottom width is used, it is important to be consistent so that W 1 and W 2 refer to either bottom widths or topwidths. 5. The average width of the bridge opening (W 2) is normally taken as the bottom width, with the width of the piers subtracted. 6. Laursen's equation will overestimate the depth of scour at the bridge if the bridge is located at the upstream end of a natural contraction or if the contraction is the result of the bridge abutments and piers. At this time, however, it is the best equation available. 7. In sand channel streams where the contraction scour hole is filled in on the falling stage, the y0 depth may be approximated by y1. Sketches or surveys through the bridge can help in determining the existing bed elevation. 8. Coarse sediments in the bed material which armor the bed may limit scour depths with live-bed contraction scour. Where coarse sediments are present, it is recommended that scour depths be calculated for live-bed scour conditions using the clear-water scour equation (given in the next section) in addition to the live-bed equation, and that the smaller calculated scour depth be used. Clear-water Contraction Scour. The recommended clear-water contraction scour equation is based on a development suggested by Laursen (1963). Its development is presented in Chapter 3 as part of the development of the critical velocity equation. The equation is:

é K Q2 ù y 2 = ê 2 /u3 2 ú ê ëDm W ú û

3/7

(7.4)

y s = y 2 − y o = (average scour depth, m)

(7.5)

where: y2 Q Dm D50 W yo K = = = = = = = Average depth in the contracted section after contraction scour, m (ft) Discharge through the bridge or on the set-back overbank area at the bridge associated with the width W, m3/s (ft3/s) Diameter of the smallest nontransportable particle in the bed material (1.25 D50) in the contracted section, m (ft) Median diameter of bed material, m (ft) Bottom width of the contracted section less pier widths, m (ft) Existing depth in the contracted section before scour, m (ft) Coefficient derived in Chapter 3 as an extension of critical velocity

7.11

Ku Ku

= =

0.025 SI units 0.0077 English units

For stratified bed material the depth of scour can be determined by using the clear-water scour equation sequentially with successive Dm of the bed material layers. 7.6.2 Computer Models for General Scour The above equations give satisfactory but conservative contraction scour depths. Computer models, if carefully used by competent hydraulic engineers experienced in their use, will give a more precise determination of contraction scour. These models can also be used for other general scour depth determinations. The one-dimensional models presently being used and maintained are BRI-STARS (Molinas 1990, 2000) and HEC-6 (U.S. Army Corps of Engineers 1993). Also, the one-dimensional water surface profile models HEC-RAS (U.S. Army Corps of Engineers 2001) and WSPRO (Arneson and Sherman 1998) and two-dimensional models FESWMS-2D (Froehlich 1996) and RMA-2V (Thomas and McAnally 1985; U.S. Army Corps of Engineers 2001) can be used to obtain the input variables for the contraction scour equations. 7.7 LOCAL SCOUR AT PIERS 7.7.1 Introduction Local scour at piers is a function of bed material size, flow characteristics, fluid properties and the geometry of the pier. The subject has been studied extensively in the laboratory since the research of Dr. Laursen in the late 1940s and 1950s (Laursen 1958, 1960, 1963; Laursen and Toch 1956; Richardson and Lagasse 1999). Richardson (1999) gives a brief listing of scour investigations in the United States. As a result of the many studies there are many equations. In general, the equations are for live-bed scour in cohesionless sand bed streams, and they give widely varying results. Since 1988, through the efforts of the USGS and FHWA, a considerable number of field measurements of local pier scour depths have been collected (Landers and Mueller 1999). The data is given by Richardson and Lagasse (1999), page 585. In this section, we give two equations for determining the ultimate local pier scour and an equation to determine the topwidth of a local pier scour hole. These equations are as follows: 1. Colorado State University's (CSU) equation. (Richardson et al. 1975) 2. FHWA’s HEC-18 equation (Richardson and Davis 2001) 3. Topwidth equation (Richardson and Abed 1999; Richardson and Davis 2001) This discussion of the equations is only for simpler flow conditions and pier geometry. Equations and methods to determine local scour depths for piers in more complex flow conditions (for example tidal flow) and pier geometry (for example a pier on piles with the pile cap at the water surface) are given in HEC-18 (Richardson and Davis 2001). The HEC-18 equation is a modification of the CSU equation resulting from additional research and field measurements that have occurred since 1975. As explained in Section 7.7.2, FHWA’s HEC-18 equation is recommended for determining the ultimate scour depth for both live-bed and clear-water scour. Briaud et al. (1999) present a method for determining local pier scour depth in cohesive bed material is given for those special occasions when it is assumed that the ultimate scour depth is not needed. For

7.12

3 was used (Fr = 0. Figure 7. The data contained 384 measurements of scour at 56 bridges. m (ft) Flow depth just upstream of the pier.3 for purposes of comparing commonly used scour equations.6) where: ys y1 K1 K2 a Fr1 V1 = = = = = = = Scour depth.3 Colorado State University's Equation The Colorado State University's Equation (Richardson et al. 7. However. the large sizes in the bed material will at some depth of scour limit the scour depth. However. Froehlich’s. a bridge with foundations in clay that is to be replaced in a few years. Mueller. As can be seen from Figure 7.5 Velocity upstream of pier. and Mueller’s modified HEC-18 equation (HEC-18BM). His comparison of these equations is given in Figures 7. which is applied to the CSU equation.3 and 7.7.8 Pier width.1 and Figure 7. do not include velocity.4 do not take into account the possibility that larger sizes in the bed material could armor the scour hole. Mueller (1996) compared 22 scour equations using field data collected by the USGS (Landers and Mueller 1996. Landers. m (ft) Froude number = V1 / (gy1) 0. some equations.5 are Shen’s. and Richardson 1999). In Figure 7. the equations are compared with some field data measurements. 1975) is as follows: ys = 2.13 . 7. The CSU equation includes the velocity of the flow just upstream of the pier by including the Froude Number in the equation.example.3) in Figure 7. that decreases the scour depth when the bed material has large particles. (1999) point out the ultimate scour in cohesive material is as deep as in sand bed material. Some of the equations have velocity as a variable (normally in the form of a Froude number). Melville and Sutherland’s.43 (7.4.2 Comparison of Pier Scour Equations Jones (1983) compared many of the more common equations.4. That is. A Froude number of 0. Laursen’s.5 gives six of his 22 comparisons. HEC18. as Briaud et al. FHWA’s HEC-18 scour depth equation has a coefficient. 0 K 1 K 2 y1 éaù ê ú ë y1 û 0. The six equations in Figure 7.2 and Equation 7. However.6 Correction for flow angle of attack of flow from Table 7.65 Fr10. The equations illustrated in Figures 7. but gives lower values of scour than the other equations.7. m/s (ft/s) 7. He concluded that the HEC-18 equation was good for design because it rarely under predicted measured scour depth.3 and 7. m (ft) Correction for pier shape from Table 7. it frequently over predicted the observed scour. such as Laursen's.4 the CSU equation encloses all the data points.

7.4. Comparison of scour formulas with field scour measurements (Jones 1983).14 . Figure 7.Figure 7.3 Comparison of scour formulas for variable depth ratios (y/a) (Jones 1983).

Comparison of pier scour equations with field measurements (after Mueller 1996). 7.Figure 7.5.15 .

m (ft) Froude Number directly upstream of the pier = V1/(gy1)1/2 Mean velocity of flow directly upstream of the pier.3 Correction factor for armoring by bed material size from Equation 7.43 (7.8) If L/a is larger than 12.0 times the pier width (a) for Fr > 0.2 ft/s2) For round nose piers aligned with the flow the depth of scour has the following limits.65 Fr10.6 Correction for flow angle of attack of flow from Table 7.7.2 and Equation 7.8 ys ≤ 3.2.65 (7. 0 K 1 K 2 K 3 K 4 K 5 ê ú y1 ë y1 û 0. 7. is recommended for both live-bed and clear-water scour.13 or 7.4 FHWA HEC-18 Equation The FHWA HEC-18 equation (Richardson and Davis 2001) to predict local scour depths at a pier.81 m/s2) (32.2 can be calculated using the following equation: L æ ö K 2 = ç Cosθ + Sinθ ÷ a è ø . m/s (ft/s) Acceleration of gravity (9.1 and Figure 7. based on the CSU equation. ys ≤ 2. The equation is: éaù ys = 2. m (ft) Flow depth just upstream of the pier.9 Correction factor for pier width from Equation 7. use L/a = 12 as a maximum in Equation 7.14 Length of pier. The equation predicts maximum pier scour depths.8 and Table 7.8 Correction factor for bed condition from Table 7.7) where: ys y1 K1 K2 K3 K4 K5 L Fr1 V1 g = = = = = = = = = = Scour depth. m (ft) Correction for pier shape from Table 7.16 .8 The correction factor for angle of attack of the flow K2 given in Table 7.4 times the pier width (a) for Fr ≤ 0.7.

For greater angles.0 2.75 3. K1.0 2. Increase in Equilibrium Pier Scour Depths.1 Medium Dunes 9> H > 3 1.0.0 1.0 Angle = skew angle of flow L = length of pier. for Angle of Attack. K3. θ.Figure 7. K2 dominates and K1 should be considered as 1.0 1. Table 7.5 1.2 to 1. Correction Factor. If L/a is larger than 12.0 2. K2. 7.3 2.1.17 .6. Common pier shapes.3 Notes: 1. Bed Condition Dune Height m K3 Clear-Water Scour N/A 1.5 3. of the Flow.9 1.1 for angles of attack up to 5 degrees.1 1.3 5. Shape of Pier Nose K1 (a) Square nose (b) Round nose (c) Circular cylinder (d) Group of cylinders (e) Sharp nose 1.1 Plane bed and Antidune flow N/A 1.2 and Equation 7. The correction factor K1 for pier nose shape should be determined using Table 7. for Bed Condition.6 1.1 Large Dunes H>9 1. use the values for L/a = 12 as a maximum in Table 7.0 1. Correction Factor.5 2.9 Table 7.for Pier Nose Shape.0 2.2. Angle L/a=4 L/a=8 L/a=12 0 15 30 45 90 1.3.8.3 3. m Table 7.0 0.1 Small Dunes 3> H > 0.5 4.

3. judgment must be exercised to reduce the value of the K2 factor by selecting the effective length of the pier actually subjected to the angle of attack of the flow. the maximum pier scour may be 30 percent greater than the predicted equation value.0 mm and D95 equal to or larger than 20 mm. Mueller and Jones (1999) developed a K4 correction coefficient from a study of 384 field measurements of scour at 56 bridges.10) and VicDx = Approach velocity corresponding to critical velocity (m/s or ft/sec) for incipient scour in the accelerated flow region at the pier for the grain size Dx (m or ft) 7. the dunes will be smaller and the maximum scour may be only 10 to 20 percent larger than.2. exists at a site during flood flow. In the unusual situation where a dune bed configuration. which is typical of most bridge sites for the flood frequencies employed in scour design. if D50 > = 2 mm and D95 > = 20 mm then K 4 = 0. The correction factor results from research by Molinas et al. For smaller streams that have a dune bed configuration at flood flow. Molinas’ research for FHWA showed that when the approach velocity (V1) is less than the critical velocity (Vc90) of the D90 size of the bed material and there is a gradation in sizes in the bed material.6. This may occur on very large rivers.15 (7. Correction Factor for Bed Material Size The correction factor K4 decreases scour depths for armoring of the scour hole for bed materials that have a D50 equal to or larger than 2. (1998) and Mueller (1996). Use of this factor will result in a significant over-prediction of scour if (1) a portion of the pier is shielded from the direct impingement of the flow by an abutment or another pier.4 ( VR ) 0. the D90 will limit the scour depth.18 . The values of the correction factor K2 should be applied only when the field conditions are such that the entire length of the pier is subjected to the angle of attack of the flow. such as the Mississippi. For such cases.9) where: VR = V1 − VicD50 VcD50 − VicD95 >0 (7. or (2) an abutment or another pier redirects the flow in a direction parallel to the pier. The correction factor K3 results from the fact that for plane-bed conditions. the maximum scour may be 10 percent greater than computed with Equation 7. with large dunes.

Engineering judgment should take into consideration the volume of traffic.7. The maximum depth of scour at a pier will often be reached during one 7.053 VcD x (7. including the CSU equation.8).7 for wide piers in shallow flow. the importance of the highway.58 ç ÷ èaø æ yö K 5 = 1. The correction factor to be applied when the ratio of depth of flow (y) to pier width (a) is less than 0.19 English units Although this K4 provides a good fit with the field data.0 mm and D95 > 20.65 for V / Vc < 1 (7.19 .25 SI units 6.14) Engineering judgment should be used in applying K5 because it is based on limited data from flume experiments. m (ft) 11. Correction Factor for Wide Piers Flume studies on scour depths at wide piers in shallow flows and field observations of scour depths at bascule piers in shallow flows indicate that existing equations.13 F 0. and the Froude Number of the flow is subcritical.11) VcDx = Critical velocity (m/s or ft/s) for incipient motion for the grain size Dx (m or ft) (7. Johnson and Torrico (1994) suggest the following equations for a K factor to be used to correct Equation 7.645 ç x ÷ è a ø 0. m (ft) Velocity of the approach flow just upstream of the pier.8 (y/a < 0.25 for V / Vc > 1 (7. excluding local scour. overestimate scour depths. cost of a failure (potential loss of lives and dollars) and the change in cost that would occur if the K5 factor is used. æyö K 5 = 2.VicD x æD ö = 0.0 mm.34 F 0. For field data an increase in D95 was always accompanied with an increase in D50. the ratio of pier width (a) to the median diameter of the bed material (D50) is greater than 50 (a/D50 > 50).13) 0.4 and VR must be greater than 0.0 ç ÷ èaø 0. 7. The minimum value of K4 is 0. the value of K4 increases rather than decreases. The bed material size must have D50 > 2. the velocity ratio terms are so formed that if D50 is held constant and D95 increases. m/s (ft/s) Grain size for which x percent of the bed material is finer.5 Pier Scour in Cohesive Bed Material The rate of scour in cohesive silt and clay bed materials is many times slower than for noncohesive materials.12) /6 /3 VcDx = K u y 1 D1 1 x y1 V1 Dx Ku Ku = = = = = Depth of flow just upstream of the pier.

if K = 0. Whereas. W = y s (K + cot θ) (7. "the maximum depth of scour in sand and in clay appears to be the same. Therefore.07 to 1.8 ys and will depend on the bottom width of the scour hole and composition of the bed material. a topwidth of 2. bonds.runoff event in a sand bed material stream. Thus. multiple columns skewed to the flow.0 to 2. and piers subjected to pressure flow.07 to 2. In water.7. However.7 Topwidth of Scour Holes The topwidth of a scour hole in cohesionless bed material from one side of a pier or footing can be estimated from the following equation (Richardson and Abed 1999). the deeper the scour hole. debris on piers. propose a method and equations to use this measurement to predict depth of scour corresponding to the duration of the flood or the design life of the bridge. it is very difficult to predict the critical shear stress for clays empirically on the basis of a few index properties.000 times faster than clay. Other researchers and field experience have also demonstrated that the scour rate in clay is many times slower than in sand. Using an Erosion Function Apparatus (EFA) to measure scour in cohesive soils Briaud et al. They note "because of the number and complexity of these bonds. the angle of repose of cohesionless material is less than the values given for air. The EFA was developed to measure the erosion rate directly. m (ft) Scour depth.7. composite pier configurations (pier on pile cap on piles exposed to the flow). At the other extreme. the topwidth would vary from 1. and Debris HEC-18 (Richardson and Davis 2001) gives procedures and equations to determine local pier scour depths for piers with exposed footings and piles groups. Briaud et al. 7.20 . flume studies of pier scour in clay bed material. therefore. the smaller the bottom width. this is confirmed by the fact that the HEC-18 equation developed from sand experiments fits this data on clay quite well. m (ft) Bottom width of the scour hole as a fraction of scour depth Angle of repose of the bed material ranging from about 30° to 44° The angle of response of cohesiveness material in air ranges from about 30° to 44°. Briaud et al. the topwidth could range from 1. Also. if the bottom width of the scour hole is equal to the depth of scour ys (K = 1). 7.7). the topwidth in cohesionless sand would vary from 2. describe the phenomena.15) where: W ys K θ = = = = Topwidth of the scour hole from each side of the pier or footing.0 ys is suggested for practical applications (Figure 7.8 ys. (1999) point out. 7. as Briaud et al. (1999) demonstrated that the scour rate for sand was approximately 1. the maximum scour depth occurred behind the pier.80 ys. Flow Conditions.6 Pier Scour for Other Pier Geometry. In addition. it will take many runoff events to get the maximum depth scour at a pier in a stream with clay bed material. in their flume studies of scour at a circular cylinder in clay bed material. and factors that give and affect cohesion in clays." This data refers to the Briaud et al. In general." They propose that the critical shear stress be measured for the clay bed material at the bridge site directly.

8 LOCAL SCOUR AT ABUTMENTS 7. This will usually give the maximum scour depth. in the model live-bed transport conditions may be ripples or dunes. 7. see Chapter 5 and HEC-23 (Lagasse et al. clear-water scour conditions).7. at flood flows in sand bed streams the sediment transport conditions will be live-bed and the bed configuration will be plane bed. the Froude number for the model should be the same as for the prototype.Figure 7. The scale between model and prototype is based on the Froude criteria. Topwidth of scour hole (HEC-18). but a careful study of the results need to be made by persons with field and model scour experience. 2001). Also.7. that is. Therefore. 7.21 .1 Introduction The two causes of scour when the flow is obstructed by the abutment and approach highway embankment are: (1) A horizontal vortex starting at the upstream end of the abutment and running along the toe of the abutment. Whereas. These are incomparable pier scour conditions.8).e. and (2) a vertical wake vortex at the downstream end of the abutment (Figure 7.8. In general. it is recommended that a bed material be used that has a critical velocity just below the model velocity (i. For additional discussion of the use of physical modeling in hydraulic design. it is not possible to scale the bed material size..8 Physical Model Studies For unusual or complex pier foundation configurations a physical model study should be made. 7.

These equations have had limited testing of computed vs. engineering judgment is required in designing foundations for abutments. measured abutment scour and will not be given. Kouchakzadeh and Townsend (1999) and Trivino and Richardson (2000) present equations based on momentum transfer for the determination of abutment scour depths.2. The abutment should be protected from local scour from both 7. and the vertical vortex that forms at the downstream end is similar to the wake vortex that forms downstream of a pier.8.22 . The wake vortex at the downstream end of the abutment also causes abutment failures.Figure 7. 2001b). In fact. Only since 1992 has research been conducted to develop equations for the field case. abutment foundations should be designed assuming no ground support (lateral or vertical) as a result of soil loss from long-term degradation and contraction scour. Methods to protect abutments and the embankments against wake vortex erosion are given in HEC-23 (Lagasse et al. The horizontal vortex at the toe of the abutment is very similar to the horseshoe vortex that forms at piers. many of the equations. Maryland SHA has developed an abutment scour program titled ABSCOUR (Chang and Davis 1999) based on Laursen’s long contraction scour equations. From this research. very little field data exist to verify the equations. b) developed equations based on his research at Georgia Institute of Technology. Research and the development of methods to determine the erosion from the wake vortex has not been conducted. In summary.8. Schematic representation of abutment scour. or that forms downstream of any flow separation. as explained in Section 7. A large amount of laboratory research has been conducted to determine the depth and location of the scour hole that develops for the horizontal (so called horseshoe) vortex that occurs at the upstream end of the abutment. are spurious correlations for the field case. Sometimes the abutment does not fail but only the approach embankment is eroded. However. numerous abutment scour equations have been developed to predict this scour depth. As a minimum. Sturm (1999a.

the discharge intercepted by the abutment is directly related to the abutment length. whereas. "The reason the equations in the literature predict excessively conservative abutment scour depths for the field situation is that. Thus. Comparison of laboratory flow characteristics to field flow conditions.23 . p. equations for predicting abutment scour would be more applicable to field condition if they included the discharge intercepted by the embankment rather than embankment length. 2001b). They stated. 457). Richardson and Richardson (1993) pointed this out in a discussion of Melville's (1992) paper and in a 1999 paper (Richardson and Lagasse 1999.2 Commentary on Abutment Scour Equations Until recently. in the laboratory flume. two equations from HEC-18 are presented for use in estimating scour depths as a guide in designing abutment foundations.8.the upstream and downstream vortexes using riprap and/or guide banks. To protect the abutment and approach roadway from scour by the wake vortex several SHAs use a 15-meter (50-ft) guide bank extending from the downstream corner of the abutment. Guidelines for the design of riprap and guide banks are given in HEC-23 (Lagasse et al. in the field. (a) (b) Figure 7. Otherwise.9. the downstream abutment and approach should be protected with riprap.9 illustrates the difference. this is rarely the case." Figure 7. the equations in the literature were developed using the abutment and roadway approach length as one of the variables. In the following sections. 7. The methods can be used for either clear-water or live-bed scour. This approach results in excessively conservative estimates of scour depth. 7.

with or without wingwalls. actually set into the channel itself. Most of the early laboratory research failed to replicate these field conditions. and alignment.3 Abutment Site Conditions Abutments can be set back from the natural stream bank. Finally.8. Richardson and Richardson (1993) noted that abutment scour depth depends on abutment shape. placed at the bankline or. In addition. and vertical wall abutments. 7. discharge intercepted by the abutment and returned to the main channel at the abutment. cross-sectional shape of the main channel at the abutment (especially the depth of flow in the main channel and depth of the overbank flow at the abutment). Orientation of embankment angle θ to the flow. the scour depth is decreased whereas the scour depth is increased for an abutment angled upstream. and shallow depths upstream of the abutment. discharge in the main channel at the abutment. sediment characteristics. Common designs include stub abutments placed on spill through slopes.Abutment scour depends on the interaction of the flow obstructed by the abutment and roadway approach and the flow in the main channel at the abutment.5. More severe abutment scour will occur when the majority of overbank flow returns to the bridge opening directly upstream of the bridge crossing.10. Less severe abutment scour will occur when overbank flows gradually return to the main channel upstream of the bridge crossing.24 . θ Figure 7. field conditions may have tree-lined or vegetated banks. An equation for adjusting abutment scour depth for embankment skew is given in Section 7. low velocities. The bridge and approach road can cross the stream and floodplain at a skew angle and this will have an effect on flow conditions at the abutment. in some cases. For an abutment angled downstream. 7. Scour at abutments can be live-bed or clear-water scour. there can be varying amounts of overbank flow intercepted by the approaches to the bridge and returned to the stream at the abutment.8. The skew angle for an abutment (embankment) is depicted in Figure 7.10. The discharge returned to the main channel at the abutment is not simply a function of the abutment and roadway length in the field case.

8. Table 7.8. As shown in Table 7. Froehlich analyzed by regression analysis.55 7.5 Froehlich’s Live-Bed Scour Equation To determine the potential depth of scour at existing bridges and to aid in the design of foundations and placement of rock riprap or guide banks at new bridges. Similarly. this coefficient will approach a value of 1. scour for vertical wall abutments with wingwalls on short abutment sections is reduced to 82 percent of the scour of vertical wall abutments without wingwalls. the effect of the spill-through slope is decreased.25 . this coefficient will approach a value of 1. (2) vertical walls without wing walls.4. the effect of the wingwall is decreased.11). For long approach road sections on the floodplain. As the length of the abutment and approach road in the floodplain increase. Figure 7.0.82 Spill-through abutment 0.4. Abutment shape.11.0. depth of scour is approximately double for vertical-wall abutments as compared with spill-through abutments for very short sections of the abutment and approach road. and (3) vertical-wall abutments with wing walls (Figure 7. These shapes can all have varying angles to the flow. As the length of the abutment and approach road in the floodplain increase. Froehlich's (1989) live-bed scour equation or an equation from HIRE (Richardson et al.00 Vertical-wall abutment with wing walls 0. Description K1 Vertical-wall abutment 1.7. Abutment Shape Coefficients for Short Abutment Sections.4 Abutment Shape The three general shapes for abutments are: (1) spill-through abutments. 170 live-bed scour measurements in laboratory flumes to obtain the following equation: 7. For long approach road sections in the floodplain. 1995) can be used.

m. m.4) Coefficient for angle of embankment to flow (θ/90)0.26 . m2. In a regression equation.10 for definition of θ) θ<90° if embankment points downstream θ>90° if embankment points upstream Length of abutment projected normal to flow. m.6 1975 and 1990 HIRE Equation An equation in HIRE was developed from Corps of Engineers field data of scour at the end of spurs in the Mississippi River (Richardson et al. m3/s. m/s. ft Note: That as L´ tends to 0. 1990).4) 7.13 (Figure 7.33 1 y1 0.é L′ ù ys = 2. ft2 Froude Number of approach flow upstream of the abutment = Ve/(gya)1/2 Qe/Ae.8. ft/s Flow obstructed by the abutment and approach embankment. ft Depth of flow at the abutment on the overbank or in the main channel. The HIRE equation is applicable when the ratio of projected abutment length (a) to the flow depth (y1) is greater than 25. ft Flow area of the approach cross section obstructed by the embankment. ft Scour depth.17) where: ys y1 Fr K1 = = = = Scour depth. 50 percent of the data are above or below the regression line. ft3/s Average depth of flow on the floodplain. This field situation closely resembles the laboratory experiments for abutment scour in that the discharge intercepted by the spurs was a function of the spur length. ys also tends to 0.61 + 1 (7. This equation can be used to estimate scour depth (y1) at an abutment where conditions are similar to the field conditions from which the equation was derived: the equation is: yS k = 4 Fr 0. 7.43 Fr 0. The 1 was added to the equation so as to encompass 98 percent of the data. m.16) where: K1 K2 K2 = = = L´ Ae Fr Ve Qe ya ys = = = = = = = Coefficient for abutment shape (Table 7. 1975. m. ft Froude Number based on the velocity and depth adjacent to and upstream of the abutment Coefficient for abutment shape (Table 7.27 K 1 K 2 ê ú ya ë ya û 0.55 (7.

In Chapter 10 of this manual. 2001a).17 for abutments skewed to the stream.17 for live-bed scour because clear-water scour equations potentially decreases scour at abutments due to the presence of coarser material. 7.9 SCOUR PROBLEMS Solved problems for contraction.16 or 7. and abutment scour in metric (SI) and English units are given in HEC-18 (Richardson and Davis 2001). use K2 for Equation 7. pier. Use Equations 7. Example problems for stream instability problems are given in HEC-20 (Lagasse et al. 7. This decrease is unsubstantiated by field data. guidelines for the design of scour and stream instability countermeasures are given in HEC-23 (Lagasse et al.27 . In addition. three design examples are given which demonstrate the use of some of the equations and methods given in this chapter. 2001b).16.To correct Equation 7. Clear-Water Scour at an Abutment.

28 .(page intentionally left blank) 7.

The types and amounts of data needed for planning and designing river crossings and lateral encroachments can vary from project to project depending upon the class of the proposed highway. and other structures. Such a report serves to guide the detailed designs. and recreational boat docks should be identified. The maps should show contours and relief. 8. and braided channels. The map may be very small scale showing towns. railroads and other highways and roads. and protection works. The site map should show detailed contours (1. The area map should be large enough to identify river systems and tributaries. and navigation channels should be clearly identified. alternative route selections and analyses of these routes should be documented in a report. diversions for irrigation and power. sand bars. There should be sufficient length of river included on the vicinity map to enable identification of stream type and to locate river meanders. Cultivated areas and urban and industrial areas.CHAPTER 8 DATA NEEDS AND DATA SOURCES The purpose of this chapter is to identify data needed for calculations and analyses which will lead to recommendations for highway crossings and encroachments of rivers. 8. preliminary calculations. in the vicinity of towns and cities should be noted on the map. The purpose of the map is to orient the highway project geographically with other area features. piers and alignments of piers. beaches. The site map is used to locate highway approach embankments. Other highways and railroads should be identified. mountain ranges. and structural designs.1 BASIC DATA NEEDS The data. Intakes for municipal and industrial water. 8. 8. picnic grounds. vegetation distribution and type.3 Site Maps Site maps are needed to determine details for hydraulic. channel changes.1. the type of river and the geographic area. cities. The direction of river flow should of course be clearly identified. Recreational areas such as camping.or 2-foot intervals).1 Area Maps An area map is needed to identify the location of the entire highway project and all streams and river crossings and encroachments involved.1.1. and provides reference background for environmental impact analysis and other needs such as application for permits and historical documentation for any litigation which may arise.2 Vicinity Maps Vicinity maps for each river crossing or lateral encroachment are needed to layout the proposed highway alignment and alternate routes. High water lines should be 8.1 . roadway.

1.1. river thalwegs. High water marks recorded photographically along with dates of occurrence are useful.5 Field Inspection A field inspection of potential highway encroachment sites of rivers should be made prior to the analysis. and specific data should be noted and briefly discussed. groundwater inflows. on which geophysical features are indicated.indicated on the site maps for the purpose of estimating flood flows and distributions across the river cross section. types of vegetation.1.4 Aerial and Other Photographs It is highly desirable in preparing vicinity and site maps that aerial photographs be obtained. Vegetation on floodplains and seasonal variations of vegetation should be recorded photographically. 8. Soil type determines the size of sediment in transport. Photographs aid the designer. Multi-image cameras use different ranges of the light spectrum to assist in identifying various features such as sewer outfalls. Land photographs (as opposed to aerial photos) of existing structures near the crossing are always helpful in documentation and evaluation of potential effects of highway construction. and groundwater flows. infiltration rates.6 Geologic Map A geologic vicinity map. Conditions of the river channel in the river reach of concern are easy to record photographically. and such pictures can be very helpful in analysis of the river reach. and glacial and river deposits which form control points on rivers are valuable in analysis of rivers. Topographic information can also be developed from aerial photographs for vicinity and site maps where such information is not readily available. is a basic need. 8. and they aid documentation. old meander channels. they are important in making detailed designs as well. Of course. Notable geologic formations should be photographed as well and supplemented with adequate notes. 2001). This has been implied in the foregoing paragraphs but is emphasized again because of the underlying importance of making first hand appraisals of specific sites before conclusions and recommendations are advanced for possible highway routes. 8. The rock formations. 8. who may not have the opportunity to visit the site. Historic aerial photographs are a valuable resource to assess river planform stability. existing bank protection works. sizes and heights of sandbars. river controls and geologic formations. to visualize crossings and encroachments. All photographs should be referenced on the site or on vicinity maps. outcroppings. and other features. Photographs of water intake works likely to be affected by the highway project should be obtained.2 . but it is not always feasible to provide opportunities for site inspection by the entire design staff. Channel geometry and roughness are important factors in river mechanics. Forms and checklists for a detailed geomorphic reconnaissance of the stream corridor are provided in HEC-20 (Lagasse et al.

Riverbed cross sections and profiles may be obtained with an ultrasonic depth sounder and are helpful in sediment transport and backwater studies.S. However. These estimates may be based on regionalized estimating procedures or other prediction models using meteorological and watershed data inputs.1. Records of the performance of existing bridges and other drainage structures should be obtained. it is necessary to estimate flood flows. For bridges which have failed. These are helpful in calculating historical flood discharges. flood control and other water resources investigation reports. as much information as possible should be obtained relative to direction of flow (angle of attack) at the piers or embankment ends. lakes and reservoirs as well as for coastal areas. 8. stages achieved by ice jams at specific locations should be noted. Historical records of damage to adjacent property and results of legal actions brought about because of damage are useful information also. Data on scour at piers of existing bridges (or at bridges which have failed) in the vicinity should be obtained. there are some streams where either a gaging station does not exist near the project site or a gaging station does not exist at all. Direct measurement of flood flows should be made when historical records may be deficient. Depth and velocity measurements need to be made at a sufficient number of subsections in a cross section to determine total flow rate. Observations of high water marks along the river reach should be made.Soil survey maps with engineering interpretations are available for a significant proportion of the United States. Temperature records are helpful in making snowmelt estimates. Each high water mark and relevant profile should be established.3 . distribution of flows. These meteorological data are available from the National Weather Service (NWS) Data Center of the National Oceanic and Atmospheric Administration (NOAA).1. Bed-material load. Information may also be obtained by direct sampling of the river. suspended load and wash load data may be obtained for some rivers in the water supply papers published by the U. and wind data are helpful in making wave height estimates on rivers. In such cases. Also. state engineers' reports. debris in the river. and estimates of average conditions can be made from rainfall data published by the NWS. 8. Geological Survey. and magnitudes of scour are useful information. 8. Discharge measurements made at various stages at a gaging site can provide data for developing a stage-discharge rating curve. sediment load data should be provided as auxiliary data for river analyses. Flood duration. They may be helpful in selecting layouts and assessing the suitability of fill materials.8 Hydraulic Data Whenever possible.7 Climatologic Data Stream gaging stations have been established on many streams throughout the United States. It is also helpful to know water temperatures.

8. When adjusting discharge records from a nearby gaging station to the project site. shape. the flood peaks are often prorated on the basis of drainage area ratios. It is also desirable to prepare a drainage map for the region upstream of the proposed highway project.1) in which n is the number of years of records. This analysis would involve hydrograph development and flow routing within the zone of influence of such highway structures. FHWA 1961). The method consists of ordering the annual peak flood discharges of record from the largest to smallest.4 . A flood-frequency curve is prepared from recorded stream flow data and augmented by estimated discharges (using Manning's equation or equivalent) from high water marks. After construction. Several methods ranging from sophisticated stochastic analysis to simple methods have been developed. Water Resources Council (1981) has adopted the log-Pearson III distribution for use as a base method for determining flood flow frequencies.5 to 0. Hydrologic data and hydraulic analyses should be documented in report form for project development. and providing background for any litigation which may arise as a consequence. the exponent of the ratio varies from 0. Slope-area calculations for 8. Approximate methods for extrapolating the range of flood-frequency curves are available but are not discussed in detail here (see HDS-2. Geological Survey in Bulletin 17B. or other) paper. particularly for events prior to stream gaging records. with delineation of size. in the event they occur. The recurrence interval.S. The annual (flood) discharge is plotted against its recurrence interval on special probability (Gumbel. A simple graphical method based on extreme value theory is reasonably satisfactory. historical floods and highwater marks. slope. The U.9 Hydrologic Data The purpose of hydrologic data is to determine the stream discharge. irrespective of chronological order. the highest flood discharge would have a recurrence interval of n + 1 years and lowest would have a recurrence interval of (1+1/n) years. It is desirable whenever possible to obtain flood histories of the river from residents and accounts by the news media. Depending on drainage basin characteristics. The greatest difficulty in constructing a flood-frequency curve is lack of sufficient and reliable data. The basic data needed are stream discharge data at the nearest gaging station. and water resource facilities such as storage reservoirs for irrigation and power and flood control projects. RI is calculated from RI = n +1 m (8. Estimates of flood discharge can be made from these accounts which are valuable in flood-frequency analysis. flood magnitudes. Sometimes a highway crossing and/or encroachment may have a significant effect on flood hydrographs and a hydrologic analysis should be made to determine the level of significance. land use.S. the documentation would be helpful in evaluating any damage from floods and failures.8.1. Thus. Details of the method and plotting paper may be obtained from the U. and m is the order (largest flood is ranked 1) of the flood magnitude. and duration and frequencies of flood prior to analysis of river behavior and design of the river encroachments and crossings.

needs to be estimated from the discussion presented in Chapters 2 and 3.S. Information regarding fishes and their river habitat should be obtainable from the state fish and game agencies. Species of trees and other vegetation should be determined. and a search through library material. 8. (FHWA 1984). the checklist should be helpful for planning a field investigation or other data acquisition program. the conveyance of the channel is calculated using the Manning equation in which the roughness coefficient. Data should also be obtained in order to enable assessment of stream turbidity during and after highway construction. 8. the reader is cautioned not to blindly accept computer printout as the final answer in estimating a flood frequency relationship. Detailed information on the location of these federal agencies across the U. interviews with local residents. the relevant types of data have been listed in Table 8. it is necessary to obtain water quality and biological data for the streams.1. Such data are not readily available for many rivers.3 DATA SOURCES The best data sources are national data centers where the principal function is to disseminate data. The list of sources in Table 8. Sometimes paleo (ancient) hydrology techniques need to be employed to resolve historic outliers at very sensitive sites.2 is provided to serve as a guide to the data collection task. n.10 Environmental Data In making environmental impact analyses of highway projects on streams and rivers. Whatever approach is used. But it might be necessary to collect data from a variety of other sources such as from a field investigation. Geological Survey. By referring to a catalog of (color) photographs. 8.2 CHECKLIST OF DATA NEEDS As an aid in collecting data preparatory to analysis of rivers and highway encroachment of rivers. Municipal water and sewage treatment facilities and industrial plants utilizing river water should have recent records regarding river water quality which will be helpful in making comprehensive environmental analyses. Wildlife information such as migration patterns of deer and elk should be determined and local game refuges should be located. For data which are not available. 8. similar channel situations to the specific site can be identified and a relatively inexperienced engineer may make a reliable estimate for n.5 .peak discharges can also be used. and some information regarding sensitivity of the flora to auto emissions should be obtained.1. Information on soil type to be used in construction of embankments would be helpful in this regard. and some judgment is required. In using this method. The data should be plotted on probability paper as analyzed by several commonly used methods. is available in Appendix A of the manual HEC-19. There may be more data items included in this table than are needed for a given project site.S. Water quality data for certain rivers can be obtained from the U.

rock outcropping dams.6 . Maps and Charts: (1) (2) (3) (4) (5) (6) Geographic Topographic Geologic Navigation Charts Potamology Surveys County and City Plats Aerial and Other Photos: (1) (2) (3) (4) (5) Large Scale Photos for Working Plans Small Scale Stereo Pairs of River and Surrounding Terrain Color Infrared Photos for Flow Patterns. restriction.Table 8. meandering. Dams. Radio. Television. and Soils: (1) (2) (3) (4) (5) (6) Discharge Records Stage-Discharge Records Flood Frequency Curves for Stations Near Site Flow Duration Curves (hydrographs) Newspaper. Checklist of Data Needs. Hydrology. Scour Zones. Bridges. Diversion. and Vegetation Ground Photos Underwater Photos Information on Existing Structures.1. Accounts of Large Floods Channel Geometry: • Main channel • Side channel • Navigation channel • Floodplain • Slopes • Backwater calculation • Bars • Sinuosity • Type (braided. straight) • Controls (falls. rapids. or Outfalls: (1) (2) (3) (4) (5) (6) (7) (8) Plans and Details Construction Details Alterations and Repairs Foundations Piers and Abutments Scour Dikes Field Investigations: • Bridge structure and repairs to bridge and approach • Damage due to ice or debris Hydraulic. diversions) Sediment Discharge: • Size distribution • Bed and Bank Material Sizes • Roughness Coefficient n Ice: • Recorded thickness • Dates of freeze up and break up • Flow patterns and jams (7) (8) 8.

and Soils (continued): (9) Regulating Structures: • Dams.Table 8. Reservoirs or Side Channel Impoundments Field Surveys: • Onsite inspections and photographs • Samples of sediments • Measure water and sediment discharge • Observe channel changes or realignment since last maps or photos • Identify high water lines or debris deposits due to recent floods • Check magnitude of velocities and direction of flow in vicinity of proposed structure • Outcroppings • Subsurface Exploration (10) (11) (12) (13) Climatological Data: (1) (2) (3) National Weather Service Records for Precipitation Wind Temperatures Land Use: (1) (2) (3) (4) (5) (6) (7) (8) (9) Zoning Maps Recent Aerial Photographs Planning Committee Records Urban Areas Industrial Areas Recreational Areas Primitive Areas Forests Vegetation 8. outfalls • Scour survey around existing piers. Hydrology. spur dikes • Inspect and photograph stabilization works. Hydraulic. Checklist of Data Needs. diversions • Intake.1. riprap sizes. abutments. Tributaries. filter blankets • Check wells for groundwater levels in areas • Install gaging stations Soils Information: • Excavation data • Borrow pits • Gravel pits • Cuts • Tunnels • Core boring logs • Well drilling logs • Soil tests • Permeability • Rock for riprap Planned and Anticipated Water Resources Projects Lakes.7 .

U. Department of Agriculture.S.State Highway Agency (4) City plans -. Geologic Division.S. Air Force. Cities Geologic Maps: (1) U. U.S.S. Commodity Stabilization Service. various State agencies. Land use capability surveys -. Geologic Survey. National Oceanic and Atmospheric Administration (2) American Society of Photogrammetry (3) Photogrammetric Engineering (4) Earth Resources Observation System (EROS). Department of The Army.S.U. commercial aerial survey and mapping firms. Apollo.city or county recorder (5) Federal reclamation project maps -. Department of the Interior.U. Department of the Interior.S.S. List of Data Sources.S. Forest Service (3) County maps -.some regular quadrangle maps show geological data also. Soil Conservation Service. Department of the Interior. Army Map Service (2) River plans and profiles -. Department of Agriculture. U. Bureau of Reclamation.commercial aerial mapping firms (6) American Society of Photogrammetry Planimetric Maps: (1) Plans of public land surveys -.8 . Department of the Interior. Topographic Division.S.) Soil Data: (1) (2) (3) (4) (5) County soil survey reports -.Surveying and Mapping Division Aerial Photographs: (1) The following agencies have aerial photographs of portions of the United States: U. Soil Conservation Service and Forest Service. Geological Survey. and State Geological Surveys Departments.S.Table 8.S.U. Department of Agriculture. Earth Resources. Topographic Division.U. Department of the Interior. National Park Service (4) Federal reclamation project maps -. Technology Satellite (ERTS) and Skylab (5) City or County Records (6) State Highway Agency Transportation Maps: (1) State Highway Agency.S. and U. Geological Survey. Department of the Interior. State Universities and State Agricultural and Conservation Agencies. (2) Large Cities Triangulation and Benchmarks: (1) (2) (3) State Engineer. Geological Survey. Department of the Interior. Land classification reports -. Bureau of Reclamation.U.U. Topographic Maps: (1) Quadrangle maps -.S.U. Department of the Interior. (Note . State Highway Agency. Photographs from Gemini. Soil Conservation Service. Hydraulic laboratory reports -. Bureau of Reclamation (5) Local areas -.U. Bureau of Land Management (2) National forest maps -.S.S. 8. Department of the Interior.S.U.U. Department of Agriculture. Conservation Division (3) National parks and monuments -. Bureau of Reclamation (6) American Society of Photogrammetry (7) ASCE Journal -.2. Department of the Interior.

Bureau of Reclamation. (4) Water Resources Publications -. Department of the Interior.U. (3) Reports -. Geological Survey. Department of Commerce.S. National Oceanic and Atmospheric Administration. Department of Commerce.S.S. National Oceanic and Atmospheric Administration. and U. (10) University Studies Sedimentation Data: (1) Water supply papers -. (2) Reports of State Engineers.2. Directory of Electric and Gas Utilities in the United States -. Bureau of Reclamation.U. Department of the Army.U. Bureau of Reclamation.U.S.U.U. Department of the Interior. Department of the Interior. Geological Survey. Public Health Service.S.U.State Public Health Departments.S. 8. Quality of Water Branch. National Oceanic and Atmospheric Administration.U. FEMA Flood Studies.U. Department of the Interior. (6) Corps of Engineers. Flood control studies. United States and Mexico. Soil Conservation Service. Bureau of Reclamation.S.S. Bureau of the Census. (4) Annual reports -. Federal Reclamation Projects -. Reports -. Hydrologic bulletin -.S.International Boundary and Water Commission.S. (8) State Highway Agency. National Oceanic and Atmospheric Administration and U. (4) Hydrometeorological reports -. Department of the Interior. (5) Cooperative study reports -. Irrigation and Drainage Data: (1) (2) (3) (4) Agricultural census reports -. and Welfare. etc. Geological Survey. Reports and Progress Reports -. Water Resources Division. Army.U.U.Federal Power Commission. Department of the Interior. Department of the Interior.U.U. Agricultural Marketing Service.U.McGraw Hill Publishing Co. Department of Commerce. Department of Commerce. (6) State Water Quality Agency. Department of the Agriculture.U. Climatological Data: (1) (2) National Weather Service Data Center. (3) Technical papers -. Power Data: (1) (2) (3) Directory of Electric Utilities -. (7) Tennessee Valley Authority. (9) USGS. Quality of Water branch.S. Department of Agriculture. (2) Reports -. (3) Geological Survey Circulars -.various power companies. (3) Annual reports -.S. Agricultural Statistics -. Department of the Interior. Bureau of Reclamation. Education.S. Quality of Water Reports: (1) Water supply papers -. Department of the Interior.Table 8.S.S. (5) Environmental Protection Agency. Department of Health. public utilities. U. List of Data Sources. (5) Hydraulic laboratory reports -. Stream Flow Data: (1) Water supply papers -. Department of Commerce. and U. Corps of Engineers. regional offices.various interstate compact commissions. State power commissions. Geological Survey. (2) Reports -.S.S. Department of the Interior.S.U.9 .S. Bureau of Reclamation.S.

The COMPENDEX database is the electronic version of the Engineering Index (Monthly/Annual). Bureau of Reclamation. and GeoRef. Soil Conservation Service. U.500 journals and selected government reports and books. which provides abstracted information from the world's significant engineering and technological literature. State Health Agencies. CIVIL ENGINEER DATABASE. Public Health Service. energy. State Water Conservation Boards or State Public Works Departments. State Departments of Public Health.4.2 ENVIRONMENTAL BIBLIOGRAPHY 1971 . Education. Department of Agriculture. State Departments of Water Resources. Education. California).000 journals. U. Inc. and Welfare.S. Bureau of Land Management. 8. 8. Department of the Army.S. TRIS. Department of the Interior.S.S.4 COMPUTERIZED LITERATURE AND DATA SEARCH Recent literature information can be retrieved from computerized databases of technical information. Hoboken. U. State Universities. Department of Health. Department of Agriculture. 8.City Water Departments. Fish and Wildlife -. Watershed Management -. More than 400 periodicals are currently indexed in ENVIRONMENTAL BIBLIOGRAPHY and the database includes more than 1. Fish and Wildlife Service. Santa Barbara..10 .Table 8. atmospheric studies. Bureau of Land Management.Present. and National Park Service. Forest Service. The principal databases related to highway and river environment are: COMPENDEX. Environmental Data: (1) (2) (3) Sanitation and public health -. water resources.U. U. State Highway Administration. Bureau of Business Research. (4) (5) 8. Basin and Project Reports and Special Reports: (1) (2) (3) (4) (5) U.1 COMPENDEX 1970 .Present.U. Department of the Interior. Department of the Interior. bimonthly updates (Environmental Studies Institute of the International Academy. Environmental Protection Agency. Bureau of Indian Affairs. Corps of Engineers. The COMPENDEX database provides worldwide coverage of approximately 3. Soil Conservation Service. and Welfare. List of Data Sources.S.S. Department of Health.U. Bureau of Mines. and nutrition and health. Fish and Wildlife Service. and planning commissions. monthly updates (Engineering Information.S. Departments of Public Works.S. New Jersey).2.4. Public Health Service. Municipal and Industrial Water Supplies -. ENVIRONMENTAL BIBLIOGRAPHY. Public Health Service. State Game and Fish Departments. The ENVIRONMENTAL BIBLIOGRAPHY covers the fields of general human ecology. power authorities. WATER RESOURCES ABSTRACTS. land resources.

C. Engineering subject areas include: Aerospace Engineering. and marketing. Transportation.Present. etc. Architectural Engineering.S. the National Asphalt Pavement Association.8.11 . the U. traffic. International information sources include the International Union of Public Transport. design.5 TRIS TRIS is the Transportation Research Information Services data base. standards. equipment. The file covers a wide range of water resource topics including water resource economics. TRIS contains more than 350. erosion. Dept. Geotechnical Engineering. Water Resources.Present. and newsletters.000 journals are scanned for TRIS. 8. management. Earthquake Engineering. Coastal. 8.S. WRA covers predominantly English-language materials and includes monographs. magazines. the U. Engineering Mechanics.). General Accounting Office. groundwater. Management.000 abstracts of completed research and summaries of research projects in progress.4. The collection is particularly strong in the literature on water planning (demand. and conference proceedings. It provides access to all the journals. universities. finance. The Civil Engineering Database is designed to provide easy bibliographic access to all American Society of Civil Engineers ASCE publications. water cycle (precipitation. Congress. Hydraulics. Bridges.4. reports patents. monthly updates (U. More than 1. of the Interior. Construction. journal articles. the U. D. construction. books. Army Corps of engineers. Port. including planning. ground and surface water hydrology. the European Conference of Ministers of Transport.S. a computerized information file maintained and operated by the Transportation Research Board (TRB) of the National Research Council. Department of Transportation's Federal Highway Administration and Federal Transit Administration. lakes. metropolitan water resources planning and management and water-related aspects of nuclear radiation and safety. cost allocations). trade and professional associations.4.S. Structural Engineering.4 WATER RESOURCES ABSTRACTS 1967 . Forensic Engineering. maintenance. Education. Department of Transportation's Federal Highway Administration. Washington. 8. Primary domestic information sources include the U. conference proceedings. manuals. Federal Transit Administration. research institutes. and the Organization for Economic Cooperation and Development.S. and Ocean Engineering. operations.3 CIVIL ENGINEERING DATABASE 1973 . Highways. Cold Regions. TRIS contains information on transportation modes and practices. Hydrology. waste treatment). the District of Columbia and Puerto Rico. and other regional and state organizations.) WATER RESOURCES ABSTRACTS is prepared from materials collected by over 50 water research centers and institutes in the United States. the highway and transportation departments of the fifty states.S. Environmental Engineering. It is sponsored by the U. Computer Practices. Irrigation and Drainage. Waterway. and the Association of American Railroads. Materials Engineering. snow. and water quality (pollution. and National Highway Traffic Safety Administration. The database covers ASCE documents published since 1973. the American Automobile Manufacturers Association. economics. Urban Planning.

S. books.6 GeoRef GeoRef is an international resource for bibliographic information about geology and geosciences. It includes more than 3.8. Geological Survey.12 . It is a comprehensive database of citations established by the American Geological Institute (AGI) in 1966.500 journals in 40 languages and references to all publications of the U. maps.8 million references to articles. reports. and theses. GeoRef provides access to over 1.4. The purpose of GeoRef is to facilitate research and development and to meet the needs of the geoscience community. 8. covering the geology of North America from 1785 to the present. conference papers.

1 . analysis and design of river crossings and encroachments. The hypothetical cases are followed by actual case histories for river crossings in the United States. medium. 9. and fundamental mechanisms and relationships of the governing physical processes must be clearly understood prior to design. In addition. In most cases. These histories document river response to highway crossings and encroachments and illustrate river response qualitatively. The principal factors to be considered in design are presented. or large. For example. 9. or it may be braided. followed by a discussion of the procedures recommended for the evaluation. or old. This chapter contains several hypothetical cases of river environments and their response to crossings and encroachments based upon geomorphic principles given in Chapter 5. a meandering river may be small.1 Types of Rivers In selecting the site for a crossing or an encroachment on a river it is necessary to give detailed consideration and study to the type of river or rivers involved. Applications of the basic principles developed in Chapters 1 through 8 are illustrated by these conceptual examples and specific case histories related to the subject matter of this manual. is necessary for a meaningful analysis of complex river response problems. Gravel and cobble bed channels are 9. The design of most complex problems in river engineering can be facilitated by a qualitative evaluation combined with a quantitative analysis. mature. The same channel can be classified as youthful.2. These factors are generally interrelated. Each of these different river types requires different design procedures. it may be essentially straight. hydrology. fluvial geomorphology. in designing training works for large sandbed channels (braided or meandering) it is unlikely that Kellner jetties alone will be useful to stabilize the bank alignment (see Chapter 6). These cases indicate the trend of change in river morphology for given initial conditions. followed by a quantitative estimate. and river mechanics to the hydraulic and environmental design of river crossings and highway encroachments. the systematic approach of a qualitative assessment of channel response.1 INTRODUCTION The objective of this chapter is to present applications of the fundamentals of hydraulics. It may be necessary to stabilize the banks with rock riprap and to control the overbank flows using jetties to achieve a set of specific purposes. This chapter uses two types of examples (conceptual and actual examples) related to river crossings and highway encroachments.CHAPTER 9 DESIGN CONSIDERATIONS FOR HIGHWAY ENCROACHMENTS AND RIVER CROSSINGS 9. A sandbed river may be meandering.2 PRINCIPAL FACTORS TO BE CONSIDERED IN DESIGN Identification of the principal factors to be considered in design of river crossings and encroachments is useful.

18 may change to a braided stream with a small increase in discharge or slope (see also Section 5. the cost of a particular crossing or encroachment can be significantly affected by its location. A meandering stream whose slope and discharge plot close to the braided river line in Figure 5. the bank material. can be used to predict this sensitivity. First. the maintenance cost required to keep the crossing functional over its estimated life. This type of river usually exists in a youthful or canyon type environment near the upper end of large river systems where the slopes are relatively steep.18. Economic considerations are equally important. Second. In addition to river form. the channel may have a sand bed and cohesive banks. a detailed survey of the characteristics of the bed and bank material coupled with river form plus other pertinent information are essential to design.3 River Characteristics The subclassifications of river form can be utilized to identify the range of conditions within which the particular river operates. several considerations are necessary. Each of these river systems behaves differently depending upon the characteristics of the floodplain material. In fact.2. It is important to know the sensitivity of any river system to change.4 River Geometry For planning a river crossing or an encroachment it is important to know the river geometry and its variation with discharge and time.5. 9. environmental factors must be considered. Figure 3. the energy gradient through the reach. it was pointed out that rivers can be essentially poised so that a small change in discharge characteristics can change a river from meandering to braided or vice versa. 9.2.normally considerably steeper than sandbed channels and in general have narrower river valleys.2 Location of the Crossing or Encroachment In selecting the site of a crossing or a longitudinal encroachment. or Chapter 3.13. 9. Depending upon the characteristics of the river and the environmental considerations.2 . and the bed material of the river both over the short term and the long term.3). unless appropriate weight is given to the environmental impacts it may not be possible to obtain permission to proceed with the project at all. In the extreme are torrential rivers. preferably. Hence. and the method of construction are some of the specific aspects that should be considered in locating the crossing. In Figure 1. It is essential to know the slope of the channel and. the beds of which are comprised of large rocks.3 of Chapter 1. relations were presented that illustrate how width and depth vary with stage at-a-section as well as along the length of a 9. It is necessary to determine if a river is relatively stable in form or is likely to be unstable. In Chapter 5. Criteria given in Chapter 5. it is important to determine other characteristics of the channel: that is. The cost of protective measures should also be considered in locating an encroachment. for example Figure 5. The length of the approaches versus the length of the crossing.2. the crossing or encroachment must mesh with the transportation system in the area. it may be formed in cobbles or it may be formed in other combinations of these materials. the cost of real estate that must be acquired to accomplish the crossing.

it is essential to develop stage-discharge relations since these relations fix key elevations of the structure in design and serve as bench-mark data when considering channel protection measures that may alter the stage of the river. as flow conditions bring about a transition from lower regime to upper regime. land uses include recreation. With this information. Similarly. Knowledge of the distribution of velocities should be coupled with a study of changes in position of the thalweg to estimate the severity of attack that may occur along the river banks and in the vicinity of the crossing. agriculture. 9. From another viewpoint.2. it is essential to know the discharge and its variation over time. urbanization. such as those due to artificial cutoffs or channelization.2. it is necessary to know the velocity distribution across the river cross-section and its variation in the river system.3 . industrial development. In the case of development. in terms of hydrologic data it is usually necessary to synthesize some of the required data. Large changes in velocity can occur in a river system with changing discharge and stage. flood control. 9. one needs to consider the vegetation cover on the watershed and the watershed response to changes in vegetation cover by human activities or by climatic changes. 9. reasonably good estimates of what can be anticipated at the site can be established. and grazing. the average velocity in the cross section may actually double. Also. From such information.7 Characteristics of the Watershed The water flowing in the river system and the sediment transported with the flow are usually intimately related to the watershed feeding the river system. geometry and land use. records of the flood flows are essential. seasonal variations in the river system can be considered. it is possible to develop simple relations showing how width and depth vary with discharge.5 Hydrologic Data It is necessary to gather all of the hydrologic data pertinent to the behavior of the river and to the design of the river crossing or encroachment. one needs to study the watershed considering its geology. may sufficiently steepen the gradient so the river operates in upper regime over its whole range of discharge. 9. Coupled with this. changes induced in the river system. As pointed out in Chapter 8. Finally. These possibilities must be considered in the detailed design. Furthermore. In synthesizing data it is very important to compare the particular watershed with other watersheds having similar characteristics. if the appropriate hydraulic and hydrologic data are available. Conventional techniques may be used to fill in missing records or it may be essential to synthesize records where limited hydrologic data exist. In a sandbed river. For most rivers. This involves determining the type of velocity distribution across the channel as well as in the vertical. Sometimes low flows may lead to a more severe local scour situation at bridge piers and footings.channel. flow duration curves can be developed. it is important to consider the low flows that the river channel will be subjected to and the possible changes in flow conditions that may be imposed on the river system as a consequence of water resources development in the area.6 Hydraulic Data At the site of a crossing or a longitudinal encroachment.2. Consequently. and design discharge values can be established depending upon the discharge frequency criteria used in the design.

The utilization of remote sensing techniques enables the skilled observer to determine which areas of the watershed are stable and which are unstable. The changing characteristics of the river with stage. Two-dimensional computer modeling. changes over time at a particular crossing affect the channel geometry. The direction of flow must be considered as a function of time. This particular topic should be studied in adequate detail so that the magnitude and intensity of the flows on the floodplain can be approximated. as discussed in Chapter 5 can provide the detailed hydraulic data to support design. One of the most common techniques to study the sources of sediment in a watershed is to employ aerial photography and remote sensing techniques coupled with ground investigations. and this must be taken into consideration. Again. intermediate and high stages.Significant changes in vegetation cover affect the amount of sediment delivered from the watershed to the river system. and the river form can have a significant effect on the intensity of attack on the approaches. the abutments. design floods usually flow in both the main channel and on the floodplain. At flood stage there is a tendency for the water to flow in the main channel in such a way as to develop chute channels across the point bars. the position of the thalweg. The characteristics of flow on the floodplain are especially relevant to the design of longitudinal encroachments. the water spills over the outside of the bends onto the floodplain. general scour and local scour. 2-dimensional computer modeling. then the information in Chapter 6 can be utilized to design against excessive contraction and local scour in order to make the highway functional with minimum maintenance over the life of the project. 9.8 Flow Alignment In order to design a safe crossing or longitudinal encroachment.9 Flow on the Floodplain Up to this point the concern has been principally with flow in the main channel. it is possible to evaluate the sediment yield as a function of time. as discussed in Chapter 5. Flow conditions on the floodplain and in the main channel can be greatly different at flood stage than at low flow. it is necessary to consider the flow alignment in detail. 9. Only by studying the characteristics and geometry of the river and the floodplain can one determine the type of flows that are apt to occur on the floodplain. Often. If the characteristics of the flow and how they vary with time are known. such as the change in velocity distribution. 9. the geometry of the crossing itself. However. can provide detailed hydraulic data on overbank and floodplain flows. the piers and embankments.2. Only with this type of information can one adequately consider the intensity of attack.4 . the duration of attack and the necessity for training works to make the river system operate within a range of conditions acceptable at the crossing or encroachment. or in some cases physical modeling. As an example.2. The detailed study of the behavior of the river over time and with varying discharge is necessary for proper design of training works. The position of the thalweg will vary with low. consider a sinuous channel. Viewing the total watershed from this perspective and using water and sediment routing techniques. Certainly.

and having determined the best site considering these important factors.2. As discussed in Chapter 5. In addition. the number and location of the piers. Similarly. existing or proposed developments up. 2001) also provides experience. the hydrologic and hydraulic characteristics of the river. the amount of allowable longitudinal encroachment into the main channel.5 . The possible necessity of holding the river in a selected alignment must also be adequately explored. present and future watershed conditions are all important inputs to the site selection. If this model is appropriately designed and utilized. it is essential to determine the need for bank stabilization. such factors as the form of the river.10 Site Selection Most of the factors cited in the preceding sections have a bearing on the final site selection. it may be important to establish the pattern of clear-water releases from a dam upstream of a crossing. For example.2. it is necessary to consider the requirements of the area to be served and the economic and environmental factors that relate to the crossing.2. variations of the river form over time. and past. and the design of training works that assist in controlling the alignment of the river relative to the crossing or longitudinal encroachment. The magnitude of local scour at the training structure must be considered. The location. 9. the principles of Chapter 6 can be applied to develop suitable designs for stabilizing the approaches and banks of the main channel. hydrologic and hydraulic data. the type of bed and bank material. it is essential to consider the shortterm response of the river system to construction. The techniques that may be utilized to investigate the short-term response at the site or in the vicinity of the crossing or encroachment involve the utilization of qualitative geomorphic relationships followed by the application of more sophisticated analyses using the principles presented in the chapters on open channel flow.11 Channel Stability Investigations In conjunction with the background information discussed in the preceding paragraphs. 23 (Lagasse et al. and design of various types of river training works must be considered. and the requirement for river training works all need to be considered.9. Having made a detailed study of possible alternate sites. In summary. The selection of training works is significantly affected by the characteristics of the river and the river system itself. and design guidance for a variety of river training works and their applicability under a range of characteristics of the river environment. the alignment of the river. it is possible to establish a mathematical model designed to route both water and sediment through the system. one can make an estimate of the extent of degradation in 9. 9. and the depth of structural support for the piers to insure against danger from local scour. Hydraulic Engineering Circular (HEC) No. The location of the longitudinal encroachment in the floodplain.12 Short-Term Response Having completed the tentative design of the crossing or the encroachment based on river form. the type and location of the abutments. it is possible to evaluate the response of the river system to both the construction of the crossing or encroachment and to other river development projects in the immediate area. sediment transport and river mechanics. selection.and downstream of the site and at the site itself should also be considered. selection. With regard to these particular issues. Knowing the type of flow the channel would be subjected to and that the water being released is clear. one can then proceed with the determination of the geometry and length of the approaches to the crossing. channel geometry.

the interrelationship of all aspects of the project. At all stages of the investigation and design. qualitative evaluation is important to determine. A generalized approach can be described.13 Long-Term Response The long-term river response at a crossing or a longitudinal encroachment and in the river system itself should be considered based on all river development projects including the highway. 9. Additionally. modification to this procedure must be made to tailor the procedure to an individual project. (2) developing several options to achieve those goals.3. The three-level procedure outlined in HEC-20 (Lagasse et al. The evaluation should begin with a qualitative assessment of the river.6).1 Approach to River Engineering Projects The evaluation and design of river crossings should proceed from a broad evaluation of the characteristics of the river and the principles to be considered in design (described in Section 9. this approach provides for back checking or "feedback" loops to insure that the interdependence of all the variables is continually adjusted (see HEC-20. Mathematical modeling can be time consuming and expensive. 9. the analysis consists of: (1) identifying the goals of the project.the channel. This type of treatment is.3 PROCEDURE FOR EVALUATION AND DESIGN OF RIVER CROSSINGS AND ENCROACHMENTS This section presents a summary of a general procedure to evaluate and design river crossings and encroachments.2.2) to detailed computations and analysis. The method begins with broad considerations and proceeds through a series of steps of increasing complexity to narrow down to the finer points of the project.1 is the recommended approach for evaluating projects in the river environment. Chapters 1 and 3) (see also Figure 7. significant advances have been made pertaining to the mathematical modeling of river systems. the instability of the banks and even the types of lateral shifting that may be induced in the river system as it affects the crossing or encroachment. 9. if possible. Due to the multi-disciplined complexity of these problems. considering both their short and long-term response. the analysis becomes more and more detailed and subsequently more quantitative. in general. it is difficult to develop a procedure which is applicable to all situations that may be encountered. Nevertheless. the amount of sediment derived from the bed and bank. (3) determining the problems and possible solutions to problems associated with each option. this approach is worth considering on important projects where determining long-term response may be critical to project success (see Section 5. requiring a substantial amount of additional data for calibration of the model. and (4) performing a qualitative assessment of all aspects of the project.1).6 . However. however. As the analysis progresses. 2001) and illustrated in Figure 9. In Level 1. beyond the scope of this particular manual. 9.

9. The following guidelines will help insure the success of the project.1. A Level 2 analysis involves a more detailed qualitative analysis combined with a quantitative evaluation. In many cases. Water can be modeled using a steady or unsteady flow rigid boundary flow model. routing both water and sediment through the study reach. A Level 3 analysis involves mathematical or physical modeling of water and sediment.2 Project Initiation The success of a project can be dramatically influenced by careful planning in the initial stages of the project. In some cases.3. Three-level analysis procedure for river engineering studies. Computation of water surface profiles using 1-dimensional computer models can be included in this level of analysis. the interrelationship between different aspects of the project and river system are adequately explained and all of the problems resolved. 9.Level 1 Simple Geomorphic Concepts and Qualitative Analysis ↓ Level 2 Advanced Geomorphic Concepts and Quantitative Engineering Analysis ↓ Level 3 Mathematical or Physical Model Studies Figure 9. Experienced modelers should be employed if sediment routing is to be performed.7 . These procedures are not always necessary. a movable boundary flow model can be employed. Sediment routing models will require a substantial historic data set and analysis time to calibrate and verify the model. the evaluation and analysis can be considered adequate at this level if the goals are met.

essential. All available data should be compiled and checked.1. it is worthwhile first of all to consider the gross effects as listed in Table 9. providing more time and funding for collection of essential data.2. Conduct a Field Reconnaissance. and design field programs to collect specific data which will be required by the analysis. Sixteen hypothetical cases are tabulated in Table 9. Collect Additional Field Data. Only the gross local. providing an interface between the field and the analysis in the office.. recommend the types of analyses which will be needed. analysis. upstream and downstream effects are identified in this table. hydraulic. When the project is conceived the goals of the project should be carefully defined and several options to meet these goals identified. Data which is unavailable or has periods of missing data should also be listed on the checklist. 9. eventually focusing on one or two options. It is also advisable that the field crews be supervised in the field by personnel who will be directly involved in the analysis and design. Field programs designed to collect the essential data could be implemented at this time. however it is recommended that a field reconnaissance and evaluation be completed prior to implementing field programs. An initial field reconnaissance should be performed by a small group of technical personnel. the field work.4 CONCEPTUAL EXAMPLES OF RIVER ENCROACHMENTS This section discusses conceptual examples of river response to highway encroachments. These options will be refined as the project progresses. geomorphic. In this way. The data checklist presented in Chapter 8 should be used as a guide. Each individual case is identified in the first column to show the physical situation that exists prior to the construction of the highway crossing. the collection of unnecessary data can be avoided.e. nonessential and optional). 2001). They should define the problems for each option and identify possible solutions to each problem. Options which are least feasible should be eliminated.1 (Level 1 analysis). The factors discussed in Section 9. 9.Develop a Project Concept. alignment and highway constraints can be identified. It is advisable that the group be multi-disciplined so that geologic. upstream.8 . hydrologic. Detailed procedures and check lists for a field reconnaissance that considers most geomorphic factors important to river engineering analyses are provided in HEC-20 (Lagasse et al. and design can be closely coordinated. Assemble Available Data. The field reconnaissance will provide a clearer definition of the project and will influence the types and quantity of additional data requirements. By designing and implementing field programs after the field reconnaissance. Missing data can be ranked according to need (i. should be considered when identifying design options. Field programs should be designed to collect only data that will be required to analyze and design the project. In an actual design situation. The field reconnaissance discussed previously is an important tool for the design and implementation of efficient field programs. In the following three columns some of the major effects (local. These personnel should be completely familiar with the types of data and the methods used to collect the data. The field reconnaissance team should identify the most favorable options. and downstream) resulting from construction of a particular crossing are given.

the local gradient of the tributary stream is significantly increased. it is possible that the depth of local scour resulting from this flow condition may be greater than the depth of local scour at high stage. 9. but have failed as a consequence of the development of greater local scour during low-flow periods.28) is one method for determining qualitative river response.1 illustrates a situation where a bridge is constructed across a tributary stream. the waterway is significantly reduced. Similar changes in the channel result if the base level is raised by some other mechanism. upstream and downstream river response. In the case illustrated for Case (1). and a decrease in flow area under the bridge. increased flood levels. Case (1) of Figure 9. Case (2) sketched in Table 9. When a river is subject to long periods of low flows. there is a tendency for the low flow to develop a new low-water thalweg in the main channel. the effect of a storage reservoir is to cause a sudden increase of base level for the upstream section of the river. Usually. If the low-water channel aligns itself with a given set of piers. if the water continues to flow under the bridge. In most cases.9 . an alluvial fan develops which in time can divert the river around the bridge. In general. changes in water and sediment input to a river often occur due to river development projects upstream from the proposed crossings or due to natural causes. The result is aggradation of the channel upstream. resulting in deposition. The effect of diversions is to decrease the river discharge downstream of the diversion with or without an overall reduction of the sediment concentration. say a tectonic uplift. The initial river conditions in Table 9. endangering the usefulness and stability of the structure. or. other river mechanics techniques can be used to predict the possibility of change in river form and to estimate the magnitude of local. Similarly. There are several documented failures where bridges have been safe in terms of local scour at high stage. The result is bank instability. It is assumed that at some point in time after the construction of the bridge the base level in the main channel has been lowered.1 involves the construction of a bridge across a tributary stream downstream of where the steeper tributary stream has reached the floodplain of the parent stream. When the base level is raised the gradient in the tributary is decreased.The Lane relation QS ~ Qs D50 (Equation 5. The average water surface elevation in the main channel acts as the base level for the tributary. similar to Case (1). Having identified the qualitative response that can be anticipated. Case (3) illustrates a situation where a bridge supported by piers and footings is constructed across a channel that is subjected to long periods of low stage.1 include storage dams or water diversions. This increased energy gradient induces head cutting and causes a significant increase in water velocities in the tributary stream. Under the new condition. streams on alluvial fans shift laterally so that the future direction of the approach flow to the bridge is uncertain. degradation downstream and a modification of the flow hydrograph. the change in gradient of the tributary stream causes significant deposition. possible major changes in the geomorphic characteristics of the tributary stream and increased local scour. These examples are used as illustrations relating to common experience. lateral channel instability.

Direction of flow at bridge site is uncertain 3 . This can result in unstable river banks and a braided streamform.Increased velocity 2 .Increase in flood stage 3 .10 .1.Large transport rate Downstream Effects 1 .Flooding 3 .Table 9.Degradation and pos sible headcutting 5 .Fan reduces waterway 2 .1.Degradation in tributary 9.Local scour 4 .Possible change of form of river 1 . increased bed material transport.Increased flood stage (2) Lowering of base level for the channel Case (4) illustrates a situation where artificial cutoffs have straightened the channel downstream of a particular crossing. Bridge Location Local Effects 1 . degradation and possible head cutting in the vicinity of the structure.Increased danger to piers due to channelization and local scour 3 .Increased bed material transport 3 .Banks unstable 6 . This causes higher velocities.Unstable channel 3 . Straightening the channel downstream of the crossing significantly increases the channel slope.Channel location is uncertain 1 .Higher velocity 3 .General scour 3 . The straightening of the main channel can drop the base level. adversely affecting tributary streams flowing into the straightened reach of the main channel.Increased transport 4 .Bank instability 5 .River may braid 7 .Erosion of banks 2 .Deposition downstream of straightened channel 2 .Steeper slope 2 . Table 9. Bridge Location Local Effects 1 .Increased transport to main channel 2 .Development of tributary bar in the main channel (1) Crossing downstream of an alluvial fan 1 .Headcutting 2 . River Response to Highway Encroachments and to River Development (continued).Loss of channel capacity 4 . River Response to Highway Encroachments and to River Development.High velocities Upstream Effects 1 .Aggradation 2 .Unstable channel 4 . which was discussed in Case (2).Bank caving 1 .At low flow a low water channel develops in river bed 2 .Aggradation 3 .Danger to bridge foundation from degradation and local scour Upstream Effects Downstream Effects ----- ----- (3) Channel characterized by prolonged low flows See local effects (4) Cutoffs downstream of crossing 1 .

Subsequently. and increased flood stages and lateral channel instability. 9. it may be difficult to achieve stability when multiple bends in a long reach are cut off. a loss of waterway at the bridge site. This contraction of the river in general increases the local velocity. It is possible to build modified reaches of main channels that do not introduce major adverse responses due to local steepening of the main channel. it is usually necessary to build a wider. the channel is too steep and bed sediments are entrained from the bed and the banks causing significant degradation. there is aggradation. In most instances. In the extreme case. Whenever the base level of a channel is raised. the depth due to general and local scour at the bridge may be significantly increased. the channel banks may become unstable due to degradation.11 . As a result. all of the sediment coming into a reservoir drops out within the reservoir. however. A chute channel can develop across the second point bar downstream. In addition. and possible changes in the characteristics of the channel are expected due to the new conditions. As stated in the preceding case. most of the sediments drop out. With sediment-free flow. Case (6) illustrates a situation where the main channel is realigned in the vicinity of the bridge crossing. there is an increase in the backwater upstream of the bridge at high flows which in turn affects other tributaries farther upstream of the crossing.Case (5) illustrates the situation where a bridge is constructed across a river immediately downstream of the confluence with a steep tributary. a pool is created extending a considerable distance upstream depending on the size of dam and slope of the channel. closing the secondary channel. A short cut off section (1 or 2 bends) can be designed to transport the same sediment loads that the river is capable of carrying upstream and downstream of the straightened reach. Such a procedure forces all of the water and sediment to pass through a reduced width. The amount depends on the magnitude of water and sediment being introduced from the tributary. the contraction can change the alignment of the flow in the vicinity of the bridge and affect the downstream channel for a considerable distance. Case (7) illustrates a bridge constructed across a main channel. the base level for the channel is raised by the construction of a dam downstream. Case (8) considers the situation where the sediment load is reduced in the channel after a bridge has been constructed. This is similar to Case (1). Also. Water released from the reservoir is mostly clear. the bridge is built across one subchannel to the island or bar formed by deposition. an island has formed in the main channel and divided flow exists. see Case (2). As the water and sediment being transported by the river encounter this pool. as in the development of storage by constructing a dam on a river. As discussed in Case (4). shallower section. increases general and local scour. increased local gradient. significant changes in river geometry. forming a delta-like structure at the mouth. may change its form. adversely affecting several meander loops downstream. In order to design a straightened channel so t hat it behaves essentially as the natural channel in terms of velocities and magnitude of bed material transport. it is possible that the degradation may cause failure of the dam and the release of a flood wave. In order to reduce the cost of the bridge structure. Upstream of the bridge. the following response at the crossing will be expected: aggradation of the bed. as its profile flattens. This may happen due to the construction of a storage dam upstream of the crossing. provides a sedimentation basin for the water flowing in the system. Also. and may increase bank instability. If the bridge is sufficiently close to the reservoir to be affected by the degradation in the channel. local velocities. If the bridge lies within the effects imposed by the new base level. and there is a possibility that the river. The tributary introduces relatively large quantities of bed material into the main channel. As a result the channel may braid. the raising of the base level of a river. A cutoff is made to straighten the main channel through the selected bridge site. local bed material transport.

Aggradation of bed 2 . River Response to Highway Encroachments and to River Development (continued).Possible destruction of structure due to dam failure 1 .Deposition of excess sediment eroded at downstream of the bridge 2 .Bank instability Upstream Effects 1 .General and local scour 4 .Contraction of the river 2 .Possible change in river form 3 .Reduced base level for tributaries.Aggradation 2 .See upstream effects (7) Raising of river base level (8) Reduction of sediment load upstream 1 .Degradation 2 .12 .1.Change in base level for tributaries 3 . Bridge Location Local Effects 1 .None if straight section is designed to transport the sediment load of the river and if it is designed to be stable when subjected to anticipated flow.Loss of waterway 3 .Backwater at flood stage 3 .Similar to local effects Table 9.Possible bank instability 5 . Otherwise same as in Case (4) 1 .Local scour 4 .Degradation 2 .Increased velocity 3 .Channel degradation 2 . River Response to Highway Encroachments and to River Development (continued). Bridge Location Local Effects 1 .See local effects 2 . increased velocity and reduced channel stability causing increased sediment transport to main channel Downstream Effects 1 .Change in river geometry 4 .Lateral channel instability Upstream Effects 1 .More severe attack at first bend downstream 3 .Table 9.Changed response of tributaries Downstream Effects 1 .Increased velocity and transport in tributaries 9.Increased flood stage 5 .Reduced flood stage 3 .Similar to local effects (5) Excess of sediment at bridge site due to upstream tributary (6) River channel relocation at crossing site 1 .Deposition in tributaries near confluences 4 .1.Possible development of a chute channel across the second point bar downstream of the bridge 1 .Aggradation causing a perched river channel to develop or changing alignment of main channel 1 .

Normally.Dam B causes aggradation 3 . Bridge Site A may aggrade due to the excess of sediment left in that tributary when clear water is diverted. and it is necessary to consider the extreme possibilities to develop a safe design. Dam B causes aggradation in the main channel (Case 7). River Response to Highway Encroachments and to River Development (continued). Upstream of Bridge B the clear water diverted from the other channel enters the storage reservoir and the water from the tributary plus the transfer water is released through a hydro-power plant. initially there may be 9. These changes in normal river flows give rise to several possible complex responses at bridge sites A and B.Final condition at bridge site is the combined effect of (1) and (2).Case (9) through Case (13) illustrate a more complicated set of circumstances.13 . These cases involve the interrelationship of Cases (1) through (8). In Case (9) the river crossing is affected by Dam A constructed upstream as well as Dam B constructed downstream. However. Situation is complex and combined interaction of dams. a diversion structure is built to divert essentially clear water by canal to the adjacent tributary on which Bridge B has been constructed.Dam A causes degradation 2 .See upstream effects In Case (10) Bridges A and B cross two major tributaries a considerable distance upstream of their confluence. It is anticipated that a larger storage reservoir may be constructed downstream of the confluence on the main stem at C. Table 9. As documented in Case 8. this analysis requires water and sediment routing techniques studying both longand short-term effects of the construction of these dams.1. Upstream of Bridge A. Bridge Location (9) Combined increase of base level and reduction of sediment load upstream Local E